PJFP.com

Pursuit of Joy, Fulfillment, and Purpose

Category: news

  • The Buzz of the Dolomites: How FPV Drones Became the Breakout Star of the 2026 Winter Olympics

    The Buzz of the Dolomites: How FPV Drones Became the Breakout Star of the 2026 Winter Olympics

    Inside the 243-gram flying cameras chasing Olympic gold at 90 mph — the pilots, the tech, the controversy, and why Milano Cortina 2026 will be remembered as the “Drone Games.”

    TLDR

    The Milano Cortina 2026 Winter Olympics have deployed FPV (First-Person View) drones as a core broadcast tool for the first time in Winter Games history. A fleet of 25 custom-built drones — weighing just 243 grams each and capable of 90 mph — are chasing bobsleds through ice canyons, diving off ski jumps alongside athletes, and orbiting snowboarders mid-trick. Built by the Dutch firm “Dutch Drone Gods” and operated by former athletes turned drone pilots, the system uses a dual-feed transmission architecture that sends ultra-low-latency video to the pilot’s goggles while simultaneously beaming broadcast-quality HD to the production truck. The result is footage that makes viewers feel like they’re sitting on the athlete’s shoulder. But the revolution comes with a buzzkill — literally. The drones’ high-pitched whine has sparked a global “angry mosquito” debate, and Italian defense contractor Leonardo has erected an invisible “Electronic Dome” of radar and jamming systems over the Dolomites to keep unauthorized drones out. Love it or hate it, FPV has graduated from experiment to Olympic standard, and the 2030 French Alps Games will inherit everything Milano Cortina pioneered.

    Key Takeaways

    • First-ever structural FPV integration at a Winter Olympics. These aren’t novelty shots — FPV is the default angle for replays and key live segments in speed disciplines at Milano Cortina 2026.
    • 25 custom drones, 15 dedicated FPV teams. The fleet is built by Dutch Drone Gods for Olympic Broadcasting Services (OBS), each unit weighing just 243 grams with top speeds of 140 kph.
    • Dual-feed transmission solves the latency problem. Pilots see 15-40ms ultra-low-latency video through their goggles while a separate HD broadcast feed with 300-400ms delay goes to the production truck via COFDM signal.
    • Pilots are former athletes. Ex-Norwegian ski jumper Jonas Sandell flies the ski jumping coverage. He anticipates the “lift” because he’s done it himself thousands of times.
    • Three-person teams modeled on military aviation. Every flight requires a pilot (goggles on, zero outside awareness), a spotter (line-of-sight, abort authority), and a director (in the OB truck, calling the live cut).
    • The inverted propeller design is the secret weapon. Mounting motors upside-down lowers the center of gravity and lets the drone “carve” air like a skier carves snow — smoother banking, cleaner footage.
    • Battery life is 5 minutes in sub-zero conditions. Heated cabins keep LiPo packs at body temperature until seconds before flight. Cold batteries can voltage-sag and drop a drone mid-chase.
    • Leonardo’s “Electronic Dome” protects the airspace. Tactical radar, RF sniffing, and electronic jamming distinguish sanctioned drones from rogue threats. Unauthorized flight is a criminal offense.
    • The “angry mosquito” controversy is real. Props spinning at 30,000+ RPM emit a 400Hz-8kHz whine that cuts through the natural soundtrack of winter sports. AI audio scrubbing is in development for 2030.
    • 93% viewership spike. 26.5 million viewers in the first five days — and FPV footage is being credited as a major factor.

    The Full Story

    As the 2026 Winter Olympic Games in Milano Cortina reach their halfway point, a singular technological narrative has emerged that eclipses even the introduction of ski mountaineering or the unprecedented decentralized venue structure spanning 400 kilometers of northern Italy. It’s not a new sport. It’s a new way of seeing sport.

    For the first time in Winter Olympics history, First-Person View drones have been deployed not as experimental novelties bolted onto the margins of production, but as the primary architectural component of the live broadcast for speed disciplines. From the icy chutes of the Cortina Sliding Centre to the vertical drops of the Stelvio Ski Centre in Bormio, a fleet of custom-engineered, high-speed micro-drones is fundamentally altering the viewer’s relationship with gravity, velocity, and fear.

    No longer tethered to fixed cable cams or zoomed-in telephoto lenses that compress depth and flatten the terror of a 90 mph descent, audiences are now riding shotgun. They’re sitting on the shoulder of a downhill skier as she threads a 2-meter gap between dolomite rock walls. They’re matching a four-man bobsled through a concrete-and-ice canyon where the walls blur into a warp-speed tunnel. They’re floating parallel to a ski jumper at the apex of a 140-meter flight, looking down at the terrifying void between athlete and earth.

    This is FPV at the Olympics. And it changes everything.

    The Hardware: 243 Grams of Purpose-Built Fury

    The drones chasing Olympic gold are nothing like the DJI Mavic sitting in your closet. They are bespoke, purpose-built broadcast machines designed to survive a hostile alpine environment while delivering cinema-grade imagery at insane speeds. The fleet comprises approximately 25 active units with 15 dedicated FPV teams, and the hardware was developed by the Netherlands-based firm Dutch Drone Gods (DDG) in partnership with Olympic Broadcasting Services.

    The engineering brief was a paradox: build something fast enough to chase a bobsled at 140 kph, yet light enough that if it ever made contact with an athlete, the damage would be survivable. The answer weighs 243 grams — just under the critical 250-gram threshold that triggers stricter aviation classification in most jurisdictions.

    Core Specs at a Glance

    FeatureSpecificationWhy It Matters
    Weight243 gramsSub-250g classification bypasses stricter aviation rules; minimizes impact energy
    Max Speed100+ kph (bursts to 140 kph / 90 mph)Matches bobsled and downhill skiing velocities
    Flight Time~5 minutes (two athlete runs)Cold degrades batteries fast; hot-swap protocol keeps packs warm until launch
    Frame DesignInverted propeller “Cinewhoop” (2.5″ to 7″)Lowered center of gravity; cleaner air over props for smoother banking
    Operating Temp-20°C to +5°CLiPo batteries pre-heated in thermal warmers to prevent voltage sag
    Pilot FeedDJI O4 Air Unit, 15-40ms latencyReflex-speed video to goggles — the pilot’s “nervous system”
    Broadcast FeedProton CAM Full HD Mini + Domo Pico Tx, 300-400ms latencyHD HDR signal via COFDM to production truck — the “visual cortex”

    The Inverted Propeller Innovation

    The single most important hardware decision DDG made was mounting the motors upside down. In a traditional drone, propellers sit above the arms and push air downward over the frame, creating turbulence. The Olympic drones flip this — motors are mounted below the arms in a “pusher” configuration.

    The physics payoff is significant. When chasing a skier through a Super-G turn, the drone must bank aggressively — sometimes 60-70 degrees. The inverted design lowers the center of gravity, allowing the drone to “carve” through the air the way a ski carves through snow. The result is footage with smooth, sweeping curves that mirror the athlete’s line rather than fighting it. And because the propellers push air away from the frame rather than washing it over the body, there’s less self-induced turbulence — critical when you’re flying centimeters from ice inside a bobsled track.

    The Dual-Feed Architecture: Two Brains, One Drone

    Here’s the fundamental problem with live FPV broadcast: a pilot flying at 90 mph needs to see what the drone sees instantly. Even a half-second delay and you’ve already crashed. But broadcast television needs high-definition, color-corrected, HDR imagery — processing that inherently introduces latency.

    The solution is elegant: each drone carries two independent transmission systems.

    The pilot feed runs through a DJI O4 Air Unit at 15-40 milliseconds of latency. It’s lower resolution, optimized purely for frame rate and response time. This is the drone’s “nervous system” — raw, twitchy, and fast. Only the pilot sees it.

    The broadcast feed uses a completely separate camera (Proton CAM Full HD Mini) and transmitter (Domo Pico Tx), running at 300-400ms latency via COFDM signal — a modulation scheme specifically chosen because it’s robust against the multipath interference caused by radio signals bouncing off dolomite rock walls and concrete sliding tracks. This feed goes straight to the Outside Broadcast van where it’s color-graded and cut into the world feed alongside 800 other cameras.

    The result: the pilot flies on instinct while the world watches in HD. Two realities, one airframe.

    The Human Element: Athletes Flying Athletes

    The most fascinating aspect of the 2026 FPV program isn’t the hardware — it’s the hiring strategy. OBS and its broadcast partners realized early on that following a ski jumper off a 140-meter hill requires more than stick skills. It requires understanding what the athlete’s body is about to do before it does it.

    So they recruited athletes.

    Jonas Sandell is a former member of the Norwegian national ski jumping team. He now flies FPV for OBS at the Predazzo Ski Jumping Stadium. His athletic background gives him something no amount of simulator time can replicate: a proprioceptive understanding of when a jumper will “pop” off the table and transition from running to flying. He anticipates the lift phase — throttling up the drone milliseconds before the visual cue — because his own body remembers the feeling. He knows the flight envelope of a ski jumper because he used to be the flight envelope.

    For the sliding sports — luge, skeleton, bobsled — the pilot known as “ShaggyFPV” from Dutch Drone Gods leads what might be the most dangerous camera crew at the Games. Flying inside the bobsled track is essentially flying inside a concrete pipe with no GPS, no stabilization assists, and a 1,500-kilogram sled bearing down at 140 kph. ShaggyFPV and his team fly up to 50 runs per session, building muscle memory of every curve and transition so deeply that the flying becomes subconscious. If a sled crashes and rides up the walls, the pilot must have a faster-than-conscious “bail out” reflex — throttle up and out of the track instantly to avoid becoming a 243-gram projectile aimed at a downed athlete.

    The Three-Person Team Protocol

    No FPV drone flies alone at the Olympics. Every unit operates under a strict three-person crew structure modeled on military aviation:

    1. The Pilot — goggles on, immersed in the FPV feed, zero awareness of the physical world. They fly on reflex and audio cues.
    2. The Spotter/Technician — maintains visual line-of-sight with the drone at all times. Monitors signal strength, battery voltage, wind, and physical hazards. Has unilateral “tap on the shoulder” authority to abort any flight, no questions asked.
    3. The Director — sits in the warmth of the OB truck, watching the drone feed alongside 20+ other camera angles. Calls the shot: “Drone 1, stand by… and TAKE.” Coordinates the cut so the drone enters the broadcast mix at exactly the right moment.

    This three-person ballet is performed hundreds of times a day across all venues. It’s the invisible choreography that makes the “wow” moments look effortless.

    The Visual Philosophy: “Movement in Sport”

    Mark Wallace, OBS Chief Content Officer, defined the visual strategy for 2026 with a two-word mandate: “Movement in Sport.” The goal isn’t just to show what happened. It’s to make the viewer feel what happened.

    In alpine skiing, the drone doesn’t just follow — it mimics. When the skier tucks, the drone drops altitude. When the skier carves, the drone banks. The camera becomes a kinesthetic mirror, conveying the violence of the vibration and the crushing G-forces in a way that a static telephoto shot from the sideline never could.

    In ski jumping, the drone tracks parallel to the athlete mid-flight, revealing the true scale of a 140-meter jump — the terrifying height, the impossible hang time, the narrow margin between textbook landing and catastrophe. Tower cameras flatten this. FPV restores it.

    In the sliding sports, the FPV drone may be the only camera capable of honestly conveying speed. Fixed trackside cameras pan so fast the sled blurs into abstraction. But the drone matches velocity, keeping the sled in razor-sharp focus while the ice walls dissolve into a warp-speed tunnel around it. For the first time, viewers at home can viscerally understand why bobsled pilots describe their sport as “controlled falling.”

    And in snowboard and freestyle at Livigno, the pilots have creative license to orbit athletes mid-trick, creating real-time “Bullet Time” effects that would have required a Hollywood rig and months of post-production just a decade ago.

    Venue by Venue: Where FPV Shines (and Struggles)

    Milano Cortina 2026 is the most geographically dispersed Olympics in history, with venues stretching across hundreds of kilometers of northern Italy. Each location presents unique challenges that force the FPV teams to adapt their hardware, techniques, and risk calculus.

    Bormio — The Vertical Wall

    The Stelvio Ski Centre hosts men’s alpine skiing on one of the steepest, iciest, most terrifying courses in the world. The north-facing slope sits in perpetual shadow. Pilots switch to heavier 7-inch drone configurations here to fight the brutal updrafts on the exposed upper mountain. The “San Pietro” jump — one of the Stelvio’s signature features — requires the drone to dive with the skier off a cliff at 140 kph, judging the athlete’s speed with centimeter-level precision. Too slow and the skier vanishes. Too fast and the shot is ruined.

    Cortina d’Ampezzo — The Amphitheater

    At the Olympia delle Tofane, women’s alpine skiing threads through massive dolomite rock formations. The challenge here is dual: RF multipath (radio signals bouncing off rock walls threaten to break up the video feed) and extreme light contrast (bright sun to deep rock shadow in seconds). The COFDM transmission system earns its keep here, and technicians in the truck ride the iris and ISO controls like a musician riding a fader.

    The Cortina Sliding Centre is the most technically demanding FPV environment at the Games. A concrete and ice canyon with no GPS signal. Pilots fly purely on muscle memory in Acro mode — no stabilization, no computer assistance, just stick and reflex. Every flight carries an abort plan because if a sled crashes, the drone needs to exit the track faster than human thought.

    Livigno — The Playground

    The open terrain of the Livigno Snow Park is where FPV gets to play. In Big Air, drones orbit rotating athletes. In Slopestyle, they chase riders across sequences of rails and jumps. When a rider checks speed to set up a trick, the drone “yaws” — turning sideways to increase drag and bleed speed instantly. It’s the most creatively expressive FPV work at the Games.

    Milan — The Indoor Frontier

    The most experimental deployment is indoors at the Mediolanum Forum for speed skating. Metal stadium beams create RF havoc, reflecting signals and causing video breakup. The solution: specialized RF repeaters and miniaturized 2.5-inch shrouded Cinewhoops safe to fly near crowds. The drones track skaters from inside the oval, revealing the tactical chess of team pursuit events in a way overhead cameras never could. Pilots fly in full manual mode with the compass disabled — the steel structure would send a magnetometer haywire.

    The Physics Problem: Flying Fast in Thin, Frozen Air

    Flying a 243-gram drone at 2,300 meters above sea level in -20°C is not the same as flying it in a parking lot in the Netherlands. The physics conspire against you at every level.

    Thin air. At the Bormio start elevation of 2,255 meters, air density is significantly lower than at sea level. Propellers generate lift by moving air, and when there’s less air to move, the props must spin faster. This draws more current, drains batteries faster, and makes the drone feel “looser” — less grip on the air, harder to hold tight lines. The DDG drones use high-pitch propellers and high-KV motors that bite aggressively into the thin atmosphere to compensate.

    Cold batteries. Lithium-polymer battery chemistry slows down as temperature drops. Internal resistance rises. When the pilot punches the throttle to chase a skier out of the start gate, the battery voltage can plummet — a phenomenon called “voltage sag” — potentially triggering a low-voltage cutoff that kills the drone mid-flight. The “Heated Cabin” protocol is not a comfort measure; it’s mission-critical. Batteries are stored at body temperature (~37°C) in thermal warmers until the final seconds before flight, and packs are swapped every two runs even if they’re not fully depleted.

    Blinding contrast. The visual environment of winter sports is an exposure nightmare: blinding white snow and ink-black shadows from rock formations. The Proton CAM was selected specifically for its HDR capability, resolving detail in both extremes simultaneously. But it’s not set-and-forget — technicians in the truck ride the exposure adjustments in real-time as the drone descends from sun to shadow and back.

    The Electronic Dome: Security in the Sky

    While OBS drones are the stars of the broadcast, they fly in one of the most securitized airspaces on the planet. The Alps present a defender’s nightmare: valleys provide radar shadows where a rogue drone can launch from a hidden floor, pop over a ridge, and be over a stadium in seconds.

    Italian defense giant Leonardo, appointed as Premium Partner for security and mission-critical communications, has erected a multi-layered Counter-UAS defense grid — an invisible “Electronic Dome” — over every venue.

    The system works in three phases:

    1. Detection. Tactical Multi-mission Radar (TMMR) — an AESA array optimized for “low, slow, and small” targets — scans the mountain clutter for anything that shouldn’t be there. Simultaneously, passive RF sensors listen for the telltale handshake signals between a remote controller and a drone.
    2. Classification. Once detected, the system must instantly determine friend or foe. OBS drones broadcast specific Remote ID signatures and operate on reserved, whitelisted frequencies coordinated with ENAC (the Italian Civil Aviation Authority). Anything detected outside the predefined 3D geofences is flagged as hostile.
    3. Mitigation. At an Olympic venue, you can’t shoot a drone down — falling debris over a crowd of thousands is not an option. Instead, Leonardo’s Falcon Shield technology performs a “soft kill,” flooding the rogue drone’s control frequencies (2.4GHz / 5.8GHz) with electronic noise. With its link severed, most consumer drones hover momentarily and then execute a Return-to-Home. Tactical teams on the ground carry handheld jamming rifles for close-range backup.

    ENAC has designated all Olympic venues as temporary “Red Zones” from February 6-22. Unauthorized drone flight in these zones isn’t a civil fine — it’s a criminal offense under the Games’ National Security designation. The US Diplomatic Security Service has gone so far as to warn American travelers that Italy will enforce strict bans and anticipates at least one “high profile drone incursion.”

    The Angry Mosquito: FPV’s Buzzing Controversy

    For all the visual brilliance, the FPV revolution has a PR problem — and it sounds like an angry insect trapped in your living room.

    Small propellers spinning at 30,000+ RPM generate a high-frequency whine in the 400Hz-8kHz range. This is precisely the frequency band where human hearing is most sensitive (we evolved to find high-pitched buzzing irritating — thanks, mosquitoes). The drone’s whine cuts through the natural soundtrack of winter sports: the roar of edges on ice, the whoosh of wind, the crunch of snow, the silence of flight. In some broadcast feeds, the drone noise overpowers everything else.

    Traditionalists argue the footage, while undeniably dynamic, can be disorienting — a “video game aesthetic” that detracts from the gravity of the Olympic moment. Others counter that the immersion is worth the acoustic cost.

    OBS CEO Yiannis Exarchos has publicly acknowledged the problem. Engineers are testing AI audio filters that can “fingerprint” the specific waveform of the DDG drone motors and subtract them from the live mix in real-time — essentially noise-canceling headphones for the broadcast. The technology isn’t fully deployed for every event in 2026, but OBS views it as a mandatory requirement for the 2030 French Alps Games.

    The Road Here: A Brief History of Olympic Drones

    Milano Cortina didn’t happen overnight. The path from aerial curiosity to broadcast infrastructure took 12 years and four Olympic cycles.

    • Sochi 2014: Drones debuted as flying tripods — slow, heavy multi-rotors capturing landscape “establishing shots” of the Caucasus Mountains. They couldn’t follow athletes and had unpredictable battery life in the Russian cold.
    • PyeongChang 2018: The 1,200-drone Intel Shooting Star light show at the Opening Ceremony was spectacular, but it was performance art, not sports coverage. Broadcast drones remained stuck on scenic B-roll.
    • Beijing 2022: COVID restrictions accelerated remote camera technology. Drones were used more aggressively in cross-country skiing and biathlon, but still as “high-eye” perspectives looking down. The latency barrier for close-proximity FPV hadn’t been cracked for broadcast-grade reliability.
    • Paris 2024: The breakthrough. OBS tested FPV in mountain biking and urban sports, proving the hybrid dual-feed transmission model worked in live production. The critical lesson: FPV pilots need to understand the sport, not just the stick. This directly shaped the athlete-recruitment strategy for 2026.
    • Milano Cortina 2026: FPV graduates from experiment to standard. It is no longer a “special feature” — it is the primary camera system for speed disciplines, treated with the same priority as a wired trackside camera on the main production switcher.

    By the Numbers

    25Active drone units across all venues
    15Dedicated FPV teams
    243gWeight of each drone (sub-250g class)
    140 kphMaximum burst speed (90 mph)
    5 minFlight time per battery in freezing conditions
    15-40msPilot feed latency (reflex-speed)
    300-400msBroadcast feed latency (HD quality)
    -20°CMinimum operating temperature
    2,300mHighest venue elevation (Tofane start)
    50 runsFlights per session for sliding sport pilots
    800Total cameras deployed across all Games coverage
    26.5MViewers in first five days (93% increase over Beijing 2022)
    12 monthsPreparation and training time per venue

    The Regulatory Stack: Why Your Drone Can’t Fly But Theirs Can

    One of the more interesting subtexts of the “Drone Games” is the dual reality playing out in Italian airspace: OBS drones are chasing bobsleds while everyone else is grounded.

    The regulatory framework operates in three layers:

    1. EU Drone Law (Commission Implementing Regulation 2019/947 and Delegated Regulation 2019/945) — defines Open, Specific, and Certified categories for all UAS operations across Europe.
    2. Italian National Implementation — ENAC and ENAV/D-Flight operationalize the rules. D-Flight provides the maps showing where you can and can’t fly, and ENAC can prohibit Open category operations inside designated UAS geographical zones.
    3. Olympic Security Overlay — temporary Red Zones and No-Fly Zones around all venue clusters from February 6-22, backed by criminal penalties under the National Security designation. These override everything else.

    OBS drones thread this needle through meticulous pre-coordination with ENAC, Italian police, venue prefectures, and the Leonardo security apparatus. Every flight path is pre-approved. Every drone broadcasts approved credentials. The “Electronic Dome” is calibrated to recognize them as friendly. A random tourist launching a Mavic? That’s a criminal act and an immediate trigger for the Counter-UAS response.

    Drone Racing: The Sport Waiting in the Wings

    There’s a fascinating meta-narrative playing out alongside the broadcast revolution: the sport of Drone Racing itself is inching toward Olympic recognition.

    Just months before the Winter Games, Drone Racing appeared as a medal event at the 2025 World Games in Chengdu. The talent overlap is striking — pilots like ShaggyFPV are already at the Olympics, just pointing their drones at athletes instead of racing gates. The FAI (World Air Sports Federation) continues to push for Olympic inclusion, and the merging of FPV broadcast culture with competitive drone culture suggests it’s a matter of when, not if.

    By the 2030s, the pilots filming the Olympics might also be competing in them.

    What Comes Next: The 2030 Legacy

    Everything pioneered at Milano Cortina — the inverted propeller design, the dual-feed transmission, the heated battery cabins, the athlete-pilot recruitment model, the three-person crew protocol — becomes the baseline standard for the 2030 Winter Games in the French Alps.

    But the technology won’t stand still. Expect further miniaturization, AI-assisted “follow-me” autonomy to reduce pilot workload, and — most critically — the perfection of real-time AI audio scrubbing to finally silence the angry mosquito without silencing the drone.

    OBS is also exploring athlete-worn microphones paired with FPV footage, which could let viewers hear the ragged breathing of a downhill skier while riding their shoulder at 90 mph. If that doesn’t make you grip your couch, nothing will.

    Thoughts

    The Milano Cortina 2026 FPV story is, at its core, a story about the collapse of distance between viewer and athlete. For decades, winter sports broadcasting has been fighting the same battle: how do you convey what it feels like to hurtle down a mountain at 90 mph to someone sitting on a couch? Telephoto lenses compress depth and kill the sense of speed. Cable cams are rigid and predictable. Helmet cams are shaky and disorienting.

    FPV cracks the problem by making the camera itself an athlete — one that flies alongside, banks with, dives with, and bleeds speed with the human it’s chasing. The footage isn’t just immersive; it’s educational. Watching an FPV shot of a downhill run, you suddenly understand why athletes describe certain sections as terrifying. You see the compression. You feel the violence of the turn. The sport makes sense in a way it never did from a static camera 200 meters away.

    The mosquito noise controversy is real but solvable — and frankly, it’s the kind of problem you want to have. It means the technology is close enough to the action to matter. AI audio scrubbing will handle it by 2030, and in the meantime, the visual revolution is worth a little buzzing.

    What’s most impressive is the human layer. The decision to hire former athletes as pilots is quietly brilliant. Jonas Sandell doesn’t just fly a drone alongside ski jumpers — he is a ski jumper who happens to be holding a transmitter instead of standing on skis. That intuitive understanding of sport physics is what separates “cool drone shot” from “footage that changes how you understand the sport.” It’s the difference between following and anticipating.

    The security dimension is equally fascinating. Leonardo’s “Electronic Dome” is essentially a small-scale military air defense system repurposed for consumer drone threats — a sign of how seriously modern event security takes the airspace layer. The fact that OBS drones need IFF-style credentialing (friend-or-foe identification, borrowed from fighter jet terminology) to avoid being jammed by their own side tells you everything about the complexity of operating sanctioned drones inside a security perimeter designed to destroy all drones.

    Looking ahead, the convergence of FPV broadcast and drone racing as a sport feels inevitable. When the pilots filming the Olympics have competition backgrounds, and the sport of drone racing is gaining World Games medals, the line between “camera operator” and “athlete” starts to blur. The FAI’s push for Olympic inclusion has never had better advertising than the footage coming out of Bormio and Cortina right now.

    Milano Cortina 2026 will be remembered as the Games where the camera stopped watching and started participating. The Buzz of Bormio may be annoying to some. But it’s the sound of sports broadcasting evolving — at 100 kilometers per hour, 243 grams at a time.

  • The Official Obsidian CLI: A Comprehensive Guide

    The Obsidian CLI allows you to control the Obsidian desktop application directly from your terminal. Whether you want to script daily backups, pipe system logs into your daily notes, or develop plugins faster, the CLI bridges the gap between your shell and your knowledge base.

    ⚠️ Early Access Warning: As of February 2026, the Obsidian CLI is in Early Access. You must be running Obsidian v1.12+ and hold a Catalyst license to use these features.

    1. Prerequisites & Installation

    Before you begin, ensure you meet the requirements:

    • Obsidian Version: v1.12.x or higher (Early Access).
    • License: Catalyst License (required for early access builds).
    • State: Obsidian must be running (the CLI connects to the active app instance).

    Setup Steps

    1. Update Obsidian: Go to Help → Check for updates. Ensure you are on the latest installer (v1.11.7+) and update to the v1.12.x early access build.
    2. Enable the CLI:
      • Open Settings → General.
      • Scroll to “Command line interface” and toggle it On.
      • Follow the prompt to “Register” the CLI. This sets up the necessary PATH variables.
    3. Restart Terminal: You must restart your terminal session for the new PATH variables to take effect.
    4. Verify: Run obsidian help. If you see a command list, you are ready.

    2. Core Concepts & Syntax

    The CLI operates in two modes: Single Command (for scripting) and Interactive TUI (for exploration).

    Interactive Mode (TUI)

    Simply type obsidian and hit enter.

    • Features: Autocomplete, command history (Up/Down arrows), and reverse search (Ctrl+R).
    • Usage: Type commands without the obsidian prefix (e.g., just daily).

    Command Structure

    The general syntax for single commands is:

    obsidian <command> [parameters] [flags]

    Parameters & Flags

    • Parameters (key=value): Quote values if they contain spaces.

      Example: obsidian create name="My Note" content="Hello World"

      Multiline: Use \n for newlines.

    • Flags: Boolean switches to change behavior.
      • --silent: Suppress output/window focusing.
      • --copy: Copy the output to the system clipboard.
      • --overwrite: Force an overwrite if a file exists.

    Targeting Vaults & Files

    • Vault Selection:
      • Default: Uses the vault in your current working directory. If not in a vault, uses the active Obsidian window.
      • Explicit: obsidian vault="My Vault" daily
    • File Selection:
      • Wikilink Style: file=Recipe (Resolves just like [[Recipe]]).
      • Exact Path: path="Folder/Subfolder/Note.md" (Relative to vault root).

    3. Essential Workflows

    Daily Notes Management

    The CLI excels at quick capture and logging without breaking your flow.

    Open Today’s Note:

    obsidian daily

    Quick Capture (Append):
    Adds text to the end of the note without opening the window.

    obsidian daily:append content="- [ ] Call Client regarding Project X" silent

    File Operations

    Create a Note:

    obsidian create name="Project Alpha" content="# Goals\n1. Launch"

    Search & Copy:
    Finds notes containing “TODO” and copies the list to your clipboard.

    obsidian search query="TODO" --copy

    Version Control

    Diff Versions:

    # Compare current file to previous version
obsidian diff file=Recipe from=1

    4. Automation & Scripting Patterns

    These patterns are ideal for shell scripts (.sh) or launchers like Alfred/Raycast.

    Pattern A: The “Inbox” Scraper

    Create a system-wide hotkey that runs this script to capture ideas instantly:

    # Appends to daily note with a timestamp
timestamp=$(date +%H:%M)
obsidian daily:append content="- $timestamp: $1" silent

    Pattern B: Automated Reporting

    Generate a file based on system data.

    # Create a note with directory listing
ls -la | obsidian create name="System Log" --stdin

    5. Troubleshooting by OS

    Windows

    Windows requires a specialized redirector because Obsidian is a GUI app.

    Fix: You may need the Obsidian.com file (available via the Catalyst Discord). Place this file alongside Obsidian.exe in your installation directory.

    macOS

    Registration usually handles this automatically. If it fails:

    Fix: Add the following to your ~/.zprofile or ~/.bash_profile:

    export PATH="$PATH:/Applications/Obsidian.app/Contents/MacOS"

    Linux

    Fix: If the symlink is missing, create it manually:

    sudo ln -s /path/to/Obsidian-AppImage /usr/local/bin/obsidian

    Command Reference Cheat Sheet

    Category Command Example Usage
    General open, search obsidian open file="Project A"
    Daily daily, daily:append obsidian daily:prepend content="Urgent!"
    Files create, move obsidian create name="Log" overwrite
    Reading read, outline obsidian read file=Recipe

    Note: Commands and syntax are subject to change during Early Access. Always rely on obsidian help within your specific build.

  • Inside X with Nikita Bier: Viral Growth, Elon Musk, and “Doing the Hard Thing”

    In a recent episode of the Out of Office podcast, Lightspeed partner Michael Mignano sat down with Nikita Bier, the Head of Product at X (formerly Twitter). Filmed in Bier’s hometown of Redondo Beach, California, the interview offers a rare, candid look into the chaotic, high-stakes world of running product at one of the world’s most influential platforms.

    Bier, famous for founding the viral apps TBH and Gas, discusses everything from his unorthodox hiring by Elon Musk to the specific growth hacks being used to revitalize a 20-year-old platform. Here is a breakdown of the conversation.

    TL;DW (Too Long; Didn’t Watch)

    • The Hire: Elon Musk hired Nikita via DM. The “interview” was a 48-hour sprint to redesign the app’s onboarding flow, which Nikita presented to Elon at 2:00 AM.
    • The Role: Bier describes his job as “customer support for 500 million people” and admits he acts as the company mascot/punching bag.
    • The Culture: X runs like a seed-stage startup. There are roughly 30 core product engineers, very few managers, and a flat hierarchy.
    • Growth Strategy: The team is focusing on “Starter Packs” to help new users find niche communities (like Peruvian politics or plumbing) rather than just general tech/news content.
    • Elon’s Management: Musk is deeply involved in engineering reviews and consistently pushes the team to “do the hard thing” rather than take shortcuts for quick growth.

    Key Takeaways

    1. Think Like an Adversary

    Bier credits his early days as a “script kiddie” hacking AOL and building phishing sites (for educational purposes, mostly) as the foundation for his product sense. He argues that understanding how to break a system is essential for building consumer products. This “adversarial” mindset helps in preventing spam, but it is also the secret to growth—understanding exactly how funnels work and how to optimize them to the extreme.

    2. The “Build in Public” Double-Edged Sword

    Nikita is a prolific poster on X, often testing feature ideas in real-time. This creates an incredibly tight feedback loop where bugs are reported seconds after launch. However, it also makes him a target. He recounted the “Crypto Twitter” incident where a critique of “GM” (Good Morning) posts led to him being meme-d as a pig for a week. The sentiment only flipped when X shipped useful features like anti-spam measures and financial charts.

    3. Fixing the Link Problem

    One of the biggest recent product changes involved how X handles external links. Historically, social platforms downrank links to keep users on-site. Bier helped design a new UI where the engagement buttons (Like, Repost) remain visible while the user reads the article in the in-app browser. This allows X to capture engagement signals on external content, meaning the algorithm can finally properly rank high-quality news and articles without penalizing creators.

    4. Identity and Verification

    To combat political misinformation without compromising free speech, X launched “Country of Origin” labels. Bier explained that this allows users to see if a political opinion is coming from a local citizen or a “grifter” farm in a different country, providing context rather than censorship.

    Detailed Summary

    From TBH to X

    The interview traces Bier’s history of building viral hits. He famously sold his app TBH (a positive polling app for teens) to Facebook, and years later, built Gas (effectively the same concept) and sold it to Discord. He dispelled the myth that he simply “sold the same app twice,” noting that while the mechanics were similar, the growth engines and social graph integrations had to be completely reinvented for a new generation.

    The Musk Methodology

    Bier provides a fascinating look at Elon Musk’s leadership style. Contrary to the idea of a distant executive, Musk conducts weekly reviews with engineers where they present their code and progress directly. Bier noted that Musk has a high tolerance for pain if it means long-term stability. For example, rewriting the entire recommendation algorithm or moving data centers in mere months—projects that would take years at Meta or Google—were executed rapidly because Musk insisted on “doing the hard thing.”

    Reviving a 20-Year-Old Platform

    The core challenge at X is growth. The app has billions of dormant accounts. Bier’s strategy relies on “resurrection”—bringing old users back by showing them that X isn’t just for news, but for specific interests. This led to the creation of Starter Packs, which curate lists of accounts for specific niches. The result has been a doubling of time spent for new users.

    The Financial Future

    Bier teased upcoming features that align with Musk’s vision of an “everything app.” This includes Smart Cashtags, which allow users to pull up real-time financial data and charts within the timeline. The long-term goal is to enable transactions directly on the platform, allowing users to buy products or tip creators seamlessly.

    Thoughts

    What stands out most in this interview is the sheer precariousness of Nikita Bier’s position. He is attempting to apply “growth hacking” principles—usually reserved for fresh, nimble startups—to a massive, entrenched legacy platform. The fact that the core engineering team is only around 30 people is staggering when compared to the thousands of engineers at Meta or TikTok.

    Bier represents a new breed of product executive: the “poster-operator.” He doesn’t hide behind corporate comms; he engages in the muddy waters of the platform he builds. While this invites toxicity (and the occasional death threat, which he mentions casually), it affords X a speed of iteration that is unmatched in the industry. If X succeeds in revitalizing its growth, it will likely be because they treated the platform not as a museum of the internet, but as a product that still needs to find product-market fit every single day.

  • Super Bowl LX (2026) By The Numbers: Production Stats, Camera Tech & Record Ad Prices

    Date: February 8, 2026
    Location: Levi’s Stadium, Santa Clara
    Matchup: Seattle Seahawks vs. New England Patriots

    As kickoff approaches, NBC, Peacock, and Telemundo are set to deliver the most technologically advanced broadcast in NFL history. Below is the breakdown of the massive production numbers defining today’s event.

    The Cost of a 30-Second Spot

    The price of airtime for Super Bowl LX has broken all previous records. NBCUniversal confirmed that inventory sold out as early as September.

    • Premium Spots: A handful of prime 30-second slots have sold for over $10 million.
    • Average Price: The average cost for a standard 30-second commercial is approximately $8 million.
    • Comparison: This is a significant jump from the $7 million average seen just two years ago.

    The Visual Arsenal: Cameras & Tech

    NBC has deployed 145 dedicated cameras. When including venue support, Sony reports over 175 total cameras are active inside the stadium.

    • Game Coverage: 81 cameras trained solely on the field.
    • Pre-Game: 64 cameras dedicated exclusively to the build-up.
    • Specialty Angles: Includes two SkyCams (one “High Sky” for tactical views) and 18 POV cameras.
    • Cinematic Style: The production is using Sony Venice 2 and Burano cinema cameras for the Halftime Show to provide a movie-like depth of field.

    The Infrastructure & Connectivity

    To connect this massive visual network, the crew has laid approximately 75 miles (396,000 feet) of fiber-optic and camera cable throughout Levi’s Stadium.

    • Audio: 130 microphones embedded around the field to capture every hit and whistle.
    • Command Center: 22 mobile production units are parked in the broadcast compound.
    • Connectivity: A massive 5G upgrade allowing for median download speeds of 1.4 Gbps for fans inside the venue.

    The Workforce & Attendance

    • Staff: Over 700 NBC Sports employees are on-site to manage the broadcast.
    • Talent: Mike Tirico (Play-by-Play), Cris Collinsworth (Analyst), Melissa Stark & Kaylee Hartung (Sideline).
    • Attendance: Expected crowd of 65,000 to 70,000 fans.

    The Entertainment Lineup

    Sources & Further Reading

  • X’s $2M+ Bet on Long-Form Writing Just Paid Off — The Internet Will Never Be the Same

    On February 3, 2026, X (@XCreators) announced the winners of its first-ever $1 Million Article Contest. The total prize pool across all winners exceeded $2.15 million.

    This special contest was a major test to see how much high-quality long-form writing could perform on the platform.

    The $1 Million Grand Prize Winner

    @beaverd – “Deloitte: A $74-Billion Cancer Metastasized Across America”
    Read the full article here (44.7 million views)

    This deeply researched piece took over 50 hours to produce. @beaverd analyzed millions of government contracts, audits, and system failures to expose how Deloitte secured $74 billion in public contracts while being linked to multiple major project failures across several states.

    • California unemployment system failures – tens of billions wasted
    • Tennessee Medicaid collapse – 250,000+ kids lost coverage
    • $1.9 billion court digitization project abandoned
    • Revolving door between Deloitte and government agencies

    Runner-Up – $500,000

    @KobeissiLetter – “President Trump’s EXACT Tariff Playbook”
    Read it here (19M+ views)

    Creator’s Choice Award – $250,000

    @thedankoe – “Full guide: how to unlock extreme focus on command”
    Read the article

    Honorable Mentions – $100,000 each

    @nickshirleyy • @wolfejosh (donating full amount to charity) • @thatsKAIZEN • @ryanhallyall

    Why This Contest Matters

    X wanted to reward serious, original long-form content. The results showed that well-researched Articles can still generate massive reach and engagement on the platform.

    What Happens Next?

    The $1 Million prize was a special one-time contest for January. However, X has stated this is “only the beginning” of their push to support high-quality long-form writing.

    With increased revenue sharing and more focus on Articles, X is clearly encouraging creators to invest in deeper, more substantial content.

    The first million-dollar Article is already live:

    https://x.com/beaverd/status/2013366996180574446

    The bar for long-form writing on X has been raised significantly.

  • The Genesis Mission: Inside the “Manhattan Project” for AI-Driven Science

    TL;DR

    On November 24, 2025, President Trump signed an Executive Order launching “The Genesis Mission.” This initiative aims to centralize federal data and high-performance computing under the Department of Energy to create a massive AI platform. Likened to the World War II Manhattan Project, its goal is to accelerate scientific discovery in critical fields like nuclear energy, biotechnology, and advanced manufacturing.

    Key Takeaways

    • The “Manhattan Project” of AI: The Administration frames this as a historic national effort comparable in urgency to the project that built the atomic bomb, aimed now at global technology dominance.
    • Department of Energy Leads: The Secretary of Energy will oversee the mission, leveraging National Labs and supercomputing infrastructure.
    • The “Platform”: A new “American Science and Security Platform” will be built to host AI agents, foundation models, and secure federal datasets.
    • Six Core Challenges: The mission initially focuses on advanced manufacturing, biotechnology, critical materials, nuclear energy, quantum information science, and semiconductors.
    • Data is the Fuel: The order prioritizes unlocking the “world’s largest collection” of federal scientific datasets to train these new AI models.

    Detailed Summary of the Executive Order

    The Executive Order, titled Launching the Genesis Mission, establishes a coordinated national effort to harness Artificial Intelligence for scientific breakthroughs. Here is how the directive breaks down:

    1. Purpose and Ambition

    The order asserts that America is currently in a race for global technology dominance in AI. To win this race, the Administration is launching the “Genesis Mission,” described as a dedicated effort to unleash a new age of AI-accelerated innovation. The explicit goal is to secure energy dominance, strengthen national security, and multiply the return on taxpayer investment in R&D.

    2. The American Science and Security Platform

    The core mechanism of this mission is the creation of the American Science and Security Platform. This infrastructure will provide:

    • Compute: Secure cloud-based AI environments and DOE national lab supercomputers.
    • AI Agents: Autonomous agents designed to test hypotheses, automate research workflows, and explore design spaces.
    • Data: Access to proprietary, federally curated, and open scientific datasets, as well as synthetic data generated by DOE resources.

    3. Timeline and Milestones

    The Secretary of Energy is on a tight schedule to operationalize this vision:

    • 90 Days: Identify all available federal computing and storage resources.
    • 120 Days: Select initial data/model assets and develop a cybersecurity plan for incorporating data from outside the federal government.
    • 270 Days: Demonstrate an “initial operating capability” of the Platform for at least one national challenge.

    4. Targeted Scientific Domains

    The mission is not open-ended; it focuses on specific high-impact areas. Within 60 days, the Secretary must submit a list of at least 20 challenges, spanning priority domains including Biotechnology, Nuclear Fission and Fusion, Quantum Information Science, and Semiconductors.

    5. Public-Private and International Collaboration

    While led by the DOE, the mission explicitly calls for bringing together “brilliant American scientists” from universities and pioneering businesses. The Secretary is tasked with developing standardized frameworks for IP ownership, licensing, and trade-secret protections to encourage private sector participation.

    Analysis and Thoughts

    “The Genesis Mission will… multiply the return on taxpayer investment into research and development.”

    The Data Sovereignty Play
    The most significant aspect of this order is the recognition of federal datasets as a strategic asset. By explicitly mentioning the “world’s largest collection of such datasets” developed over decades, the Administration is leveraging an asset that private companies cannot easily duplicate. This suggests a shift toward “Sovereign AI” where the government doesn’t just regulate AI, but builds the foundational models for science.

    Hardware over Software
    Placing this under the Department of Energy (DOE) rather than the National Science Foundation (NSF) or Commerce is a strategic signal. The DOE owns the National Labs (like Oak Ridge and Lawrence Livermore) and the world’s fastest supercomputers. This indicates the Administration views this as a heavy-infrastructure challenge—requiring massive energy and compute—rather than just a software problem.

    The “Manhattan Project” Framing
    Invoking the Manhattan Project sets an incredibly high bar. That project resulted in a singular, world-changing weapon. The Genesis Mission aims for a broader diffusion of “AI agents” to automate research. The success of this mission will depend heavily on the integration mentioned in Section 2—getting academic, private, and classified federal systems to talk to each other without compromising security.

    The Energy Component
    It is notable that nuclear fission and fusion are highlighted as specific challenges. AI is notoriously energy-hungry. By tasking the DOE with solving energy problems using AI, the mission creates a feedback loop: better AI designs better power plants, which power better AI.

  • NVIDIA (NVDA) Q3 FY2026 Earnings: $57B Record Revenue, Blackwell “Off the Charts,” $65B Guidance – The AI Boom Is Still Accelerating

    November 20, 2025 – NVIDIA just delivered the most dominant quarter in the history of tech and told the world the next one will be even bigger. The market is partying like it’s 2021.

    TL;DR

    • Revenue $57.01B (+62% YoY, beat by ~$1.8–2B)
    • Data Center $51.2B (+66% YoY, +$10B sequentially) – now 90% of total revenue
    • GAAP EPS $1.30 (+67% YoY)
    • Q4 guidance $65B (±2%) – obliterates street $61.98B (some buyside whispers were $75B → Jensen sandbagging again)
    • Blackwell sales “off the charts”, cloud GPUs completely sold out for the foreseeable future
    • CFO Colette Kress confirmed ≈$500B Blackwell + Rubin revenue visibility 2025–2026 (analysts now calling it $500B pipeline through FY2027)
    • Gross margin 73.6% (tiny miss due to Blackwell ramp costs), guided back to 75.0% next quarter
    • Free cash flow $22.1B in a single quarter
    • Top 4 customers = 61% of revenue (22% / 15% / 13% / 11%) – concentration risk is real but demand makes it a feature
    • Stock ripped +5.5% after-hours → +$220B+ market cap in minutes, lifting entire AI complex

    Key Takeaways

    • Demand is not slowing — it’s compounding. Jensen: “Compute demand keeps accelerating and compounding across training and inference — each growing exponentially. We’ve entered the virtuous cycle of AI.”
    • Blackwell ramp is unprecedented – already the majority of new Data Center mix, sold out for months, driving the entire $10B sequential jump
    • Gaming ($4.3B) and Automotive ($592M) missed estimates → literally nobody cares when Data Center grew $10B in one quarter
    • Customer concentration: Four hyperscalers = 61% of revenue. Everyone knows who they are. Everyone also knows they can’t build without NVIDIA
    • Margins dipped to 73.6% only because of Blackwell complexity/HBM costs – guided 75% next quarter, street relieved
    • Balance sheet is absurd: $60.6B cash + $22.1B quarterly FCF. Berkshire is only ~$320B ahead
    • Physical AI multi-trillion opportunity already “multi-billion” today

    Detailed Summary

    NVIDIA printed $57.01 billion in a single quarter — a number larger than the entire annual revenue of 99% of public companies. Data Center alone did $51.2 billion (+66% YoY, +$10 billion sequentially). Let that sink in.

    Blackwell is not “ramping” — it’s exploding. It is already the majority of new Data Center revenue and cloud providers are in a literal bidding war for every wafer. Jensen was blunt: “Blackwell sales are off the charts, and cloud GPUs are sold out.”

    Yes, Gaming and Automotive missed estimates (who cares), Pro Visualization crushed it (+56% YoY), but the only number that matters is the $500 billion in confirmed Blackwell + Rubin orders the company can already see through calendar 2026 (Bloomberg Intelligence now calling it $500B pipeline through fiscal 2027).

    China export restrictions? Effectively $0 impact in guidance. The rest of the planet is making up for it and then some — sovereign AI factories, enterprises, everyone is building.

    Networking (Spectrum-X + InfiniBand) up ~162% YoY to $8.2B+ — the hidden monster line item nobody talks about.

    Market & Analyst Reaction

    Initial spike was +4%, then kept climbing → closed extended trading up ~5.5%, adding north of $220 billion in market cap. Entire AI food chain ripping: CoreWeave +4%, Nebius +4%, AMD +2%, Micron +2%, Broadcom +2%, Super Micro +8%.

    Goldman Sachs (James Schneider) first note post-earnings:

    “Strong quarter with upside to guidance should provide relief for the stock… We expect the stock to trade higher following a stronger quarter and guidance relative to the Street.”

    X was pure euphoria last night – here are some of the top posts (all >5K likes):

    • https://x.com/EconomyApp/status/1991259207878996127 ← Clean chart
    • https://x.com/Quartr_App/status/1991259508941734389 ← Jensen quote card
    • https://x.com/KobeissiLetter/status/1991255966235419112 ← +$205B market cap meme
    • https://x.com/amitisinvesting/status/1991263435493974047 ← Full breakdown thread
    • https://x.com/FromValue/status/1991275128439123451 ← “NVIDIA just printed more FCF than most companies make in revenue”

    My Thoughts

    This was a “relief rally on steroids”. Anyone still waiting for the AI capex slowdown just got obliterated. The $500 billion visibility isn’t hopium — it’s what they can already see in purchase orders.

    The moat is now impenetrable: CUDA + NVLink + Spectrum-X + Grace CPU + Blackwell/Rubin roadmap = Microsoft Windows-level lock-in for the AI era.

    At ~44× forward earnings the stock looks expensive until you realize the base case is now ~$260–280B annual revenue run-rate by late 2026. That puts the multiple in the low 20s. That is no longer the bull case — that’s the new floor.

    The Christmas rally is officially back on. NVIDIA just saved it.

  • Cloudflare Down November 18 2025: Massive Global Outage Takes X (Twitter), ChatGPT, Discord, Spotify, League of Legends & Thousands of Websites Offline

    FINAL UPDATE – Post-Mortem Released: Cloudflare has released the detailed post-mortem for the November 18 event. The outage was caused by an internal software error triggered by a database permission change, not a cyberattack[cite: 25, 26]. Below is the technical breakdown of exactly what went wrong.

    TL;DR – The Summary

    • Start Time: 11:20 UTC – Significant traffic delivery failures began immediately following a database update.
    • The Root Cause: A permission change to a ClickHouse database caused a “feature file” (used for Bot Management) to double in size due to duplicate rows[cite: 26, 27, 81].
    • The Failure: The file grew beyond a hard-coded limit (200 features) in the new “FL2” proxy engine, causing the Rust-based code to crash (panic)[cite: 190, 191, 194].
    • Resolution: 17:06 UTC – All systems fully restored (Main traffic recovered by 14:30 UTC)[cite: 32, 90].

    The Technical Details: A “Panic” in the Proxy

    The outage was a classic “cascading failure” scenario. Here is the simplified chain of events from the report:

    • The Trigger (11:05 UTC): Engineers applied a permission change to a ClickHouse database cluster to improve security. This inadvertently caused a query to return duplicate rows[cite: 160, 172].
    • The Bloat: This bad data flowed into a configuration file used by the Bot Management system, causing it to exceed its expected size[cite: 27, 125].
    • The Crash: Cloudflare’s proxy software (specifically the FL2 engine written in Rust) had a memory preallocation limit of 200 features. When the bloated file hit this limit, the code triggered a panic (specifically called Result::unwrap() on an Err value), causing the service to fail with HTTP 500 errors[cite: 190, 218, 219].
    • The Confusion: To make matters worse, Cloudflare’s external Status Page also went down (returning 504 Gateway Timeouts) due to a coincidence, leading engineers to initially suspect a massive coordinated cyberattack.

    Official Timeline (UTC)

    Time (UTC) Status Event Description
    17:06 Resolved All services resolved. Remaining long-tail services restarted and full operations restored[cite: 268].
    14:30 Remediating Main impact resolved. A known-good configuration file was manually deployed; core traffic began flowing normally [cite: 32, 268].
    13:37 Identified Engineers identified the Bot Management file as the trigger and stopped the automatic propagation of the bad file [cite: 268].
    13:05 Mitigating A bypass was implemented for Workers KV and Access to route around the failing proxy engine, reducing error rates [cite: 267].
    11:20 Outage Starts Network begins experiencing significant failures to deliver core traffic .
    11:05 Trigger Database access control change deployed[cite: 267].

    Final Thoughts

    Cloudflare’s CEO Matthew Prince was direct in the post-mortem: “We know we let you down today”[cite: 37]. The company has identified the specific code path that failed and is implementing “global kill switches” for features to prevent a single configuration file from taking down the network in the future[cite: 259].

    Read the full technical post-mortem: Cloudflare Blog: 18 November 2025 Outage

  • Project NOVA Reaches Zero Power Criticality Milestone at NNSS: A Major Step Forward for Advanced Nuclear Energy

    Project NOVA Reaches Zero Power Criticality Milestone at NNSS: A Major Step Forward for Advanced Nuclear Energy

    TL;DR:

    On November 17, 2025, Valar Atomics and Los Alamos National Laboratory announced that the NOVA Core – a HALEU TRISO-fueled, graphite-moderated HTGR test assembly – successfully reached zero-power (“cold”) criticality at the National Criticality Experiments Research Center (NCERC) in Nevada. This marks the first time a venture-backed private nuclear company has ever achieved criticality, validating the physics of Valar’s upcoming Ward250 reactor and clearing a major technical de-risking milestone on the path to gigawatt-scale carbon-free power.

    Key Takeaways

    • Zero-power criticality achieved at 11:45 AM PT on November 17, 2025
    • First criticality ever achieved by a venture-funded nuclear startup
    • Conducted at the United States’ only general-purpose critical experiments facility (NCERC, Nevada National Security Site)
    • Uses the exact same HALEU TRISO fuel, graphite moderator, and reactivity control scheme as the commercial Ward250 reactor
    • Directly validates Valar Atomics’ proprietary neutronics models and simulation stack
    • Builds on the 2024 Deimos critical assembly; NOVA is the high-fidelity physics twin of Ward250
    • Clears the path for hot (powered) criticality and full-temperature testing in 2026
    • Supported by DOE’s Advanced Reactor Pilot Program (target: full criticality by July 4, 2026) and Executive Order 14301
    • Strong public endorsement of the Trump administration’s “make nuclear great again” push

    Detailed Summary of the Announcement

    On November 17, 2025, Los Alamos National Laboratory (LANL) and Valar Atomics jointly announced that the NOVA Core, operating on LANL’s Comet critical assembly machine at the National Criticality Experiments Research Center (NCERC) inside the Nevada National Security Site (NNSS), had achieved zero-power criticality at exactly 11:45 AM Pacific Time.

    Approach-to-critical experiments began on November 12, 2025, and the core went critical five days later – an impressively rapid and safe execution that highlights both Valar’s engineering maturity and NCERC’s world-class operational capability.

    What is zero-power (“cold”) criticality?
    Cold criticality is the moment when a nuclear core sustains a stable neutron chain reaction (k_eff = 1.000) without external neutron sources, but at room temperature and with essentially zero fission power (typically microwatts to a few watts). No heat is removed by coolant flow, and temperatures remain ambient. It is the nuclear equivalent of “first breath” or “first heartbeat” – proof that the fundamental physics of the core design works exactly as modeled.

    Project NOVA (Nuclear Observations of Valar Atomics) is a multi-week campaign of criticality experiments designed to:

    • Measure integral neutronics parameters (reactivity coefficients, control rod worth, burnable poison performance, etc.)
    • Validate Valar’s in-house Monte Carlo and deterministic neutronics codes
    • Provide high-fidelity benchmark data for the Ward250 reactor currently under construction in Utah

    The NOVA Core is a graphite-moderated, helium-cooled-concept test bed fueled with High-Assay Low-Enriched Uranium (HALEU) TRISO particles – the same fuel form and enrichment Valar will use commercially. Reactivity control is provided by boron-carbide elements in stainless-steel cladding, mirroring the Ward250 design.

    The central portion of the core was designed and fabricated entirely by Valar Atomics, while LANL provided the Comet universal assembly machine, reflectors, instrumentation, safety envelope, and decades of criticality-safety expertise.

    Quotes from Leadership

    • Isaiah Taylor (Founder & CEO, Valar Atomics): “Zero power criticality is a reactor’s first heartbeat, proof the physics holds… This moment marks the dawn of a new era in American nuclear engineering — one defined by speed, scale, and private-sector execution with closer federal partnership.”
    • Max Ukropina (Head of Projects): “President Trump asked industry and the labs to make nuclear great again. We got together and decided to start with the basics of fission. This team delivered incredible results safely so we can keep moving up the technical ladder.”
    • Sonat Sen (Lead Core Designer): “Project NOVA provides us with real-world data which will help us answer key questions about TRISO fuel performance in our core and validate our proprietary software stack.”

    Why This Milestone Matters – Technical & Strategic Context

    Reaching criticality in a national-lab critical facility is widely regarded as the single biggest technical de-risking event for any new reactor design. Before today, no venture-backed nuclear company had ever achieved criticality on their own core. Legacy players (NuScale, TerraPower, Kairos Power, X-energy, etc.) have either used legacy government assemblies or have not yet gone critical with their exact commercial fuel and geometry.

    Valar Atomics has now leapfrogged the field by:

    1. Using actual commercial-spec HALEU TRISO (not surrogates)
    2. Replicating the exact Ward250 moderator-to-fuel ratio and control scheme
    3. Collecting integral data months ahead of first fuel load at Ward250
    4. Demonstrating that a small private team can execute at national-lab speed and safety standards

    This positions Valar to move aggressively into hot zero-power testing, helium loop commissioning, and ultimately full-power, full-temperature operation of Ward250 in 2026 – aligning perfectly with the DOE’s goal of new reactor criticality by Independence Day 2026.

    My Thoughts & Broader Implications

    1. Speed is the new moat. From Deimos (2024) → NOVA criticality (2025) → Ward250 power operations (2026) in roughly 24 months is an absolutely blistering pace by historical nuclear standards. Valar is proving that private capital + national lab partnership + focused scope can compress decades into years.

    2. TRISO + Graphite + Helium is having its moment. The combination of walk-away-safe TRISO fuel, high-temperature capability (>750°C), and modular factory fabrication is rapidly becoming the consensus Gen-IV architecture for private deployment. NOVA just added the strongest data point yet that the neutronics actually work as advertised.

    3. National labs are back as force multipliers. NCERC’s ability to take a private core, insert it into the Comet machine, and go critical in under a week with zero safety incidents is a national strategic asset. The close LANL–Valar collaboration is exactly the model the Trump administration appears to want: labs providing capability, private sector providing speed and capital.

    4. AI + Nuclear inflection point. Valar has been explicit that their ultimate product is gigasites – clusters of thousands of HTGRs powering hyperscale data centers, hydrogen electrolysis, and desalination. Today’s criticality is concrete evidence that the energy bottleneck for the AI build-out may actually be solvable in this decade.

    5. First of many. If Valar can replicate this model – design core → validate at NCERC → deploy Ward250 → scale factory production – we are looking at a genuine nuclear renaissance led by American startups rather than slow-moving utilities or foreign state-owned entities.

    Wrap Up

    November 17, 2025, will be remembered as the day a venture-backed nuclear company first split the atom under its own design. Project NOVA’s successful cold criticality is not just a technical checkbox – it is a cultural and strategic turning point for the entire industry.

    The physics works. The team can execute. The labs are partnering at speed. The policy tailwinds are strong.

    We are witnessing the birth of the next era of American nuclear dominance – and it’s moving a lot faster than anyone predicted.