How I Hunted Down a Hidden QR Code in Jerry Rig Everything's Robot Video

TL;DR: Jerry Rig Everything released a robot video and teased a hidden QR code. I missed it after watching the video five times (yes, five—I’m apparently a master of missing the obvious), so I wrote a script to find it for me! Here’s how I did it, with all the code and results. Spoiler: it was right there the whole time. 😂

📺 The Video

He attacked me right out of the box… Are Humans Cooked?

🕵️ The Challenge

Jerry Rig Everything mentioned he hid a QR code in his latest robot video, “He attacked me right out of the box… Are Humans Cooked?” I watched it (half-assed) five times and still missed it. Time to automate! (Because why use your eyes when you can use Python?)

📝 The Plan

1. Download the video
2. Extract frames with timestamps
3. Scan each frame for QR codes using Python & OpenCV
4. Review the results and laugh at myself

1️⃣ Downloading the Video

I used yt-dlp to grab the best quality video:

./yt-dlp -f "bestvideo[ext=mp4]" https://www.youtube.com/watch?v=tF0woP73AXs

2️⃣ Extracting Frames with ffmpeg

To scan for QR codes, I needed individual frames. I used ffmpeg to extract one frame per second (every 30 frames at 30fps), overlaying the timestamp for reference:

ffmpeg -hwaccel cuda \
  -i input.mp4 \
  -vf "select='not(mod(n,30))',\
drawtext=fontfile=/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf:\
text='%{pts\\:hms}':\
x=10: y=H-th-10:\
fontsize=24: fontcolor=white: box=1: boxcolor=0x00000099" \
  -vsync vfr \
  -frame_pts 1 \
  output/frame_%d.png

3️⃣ Scanning for QR Codes with Python & OpenCV

I wrote a Python script to scan all PNG frames for QR codes in parallel, printing progress and any decoded data. Here’s the core logic:

def process_file(args):
    folder_path, filename = args
    file_path = os.path.join(folder_path, filename)
    img = cv2.imread(file_path)
    if img is None:
        return (filename, None)
    detector = cv2.QRCodeDetector()
    data, points, _ = detector.detectAndDecode(img)
    return (filename, data if points is not None and data else None)

And the script to run it across all frames:

def scan_folder_for_qr_codes(folder_path, max_workers=None):
    png_files = [f for f in os.listdir(folder_path) if f.lower().endswith('.png')]
    total = len(png_files)
    if total == 0:
        print("No PNG files found in:", folder_path)
        return {}

    results = {}
    with ProcessPoolExecutor(max_workers=max_workers) as executor:
        future_to_file = {
            executor.submit(process_file, (folder_path, fname)): fname
            for fname in png_files
        }

        processed = 0
        for future in as_completed(future_to_file):
            filename, data = future.result()
            processed += 1
            percent = processed / total * 100
            print(f"[{processed}/{total}] ({percent:5.1f}%) {filename}", end='\r')

            if data:
                print()
                print(f" → QR code detected in {filename}: {data}")
                results[filename] = data

    print()
    return results

The full script is available in the repo, but this is the gist!

4️⃣ Results

Here’s what the script found:

python3 scan_qr.py --workers 50 ./output/
[37/935] (  4.0%) frame_23130.png
 → QR code detected in frame_23130.png: [redacted-qr-link]
[63/935] (  6.7%) frame_23100.png
 → QR code detected in frame_23100.png: [redacted-qr-link]
[229/935] ( 24.5%) frame_22920.png
 → QR code detected in frame_22920.png: [redacted-qr-link]
[438/935] ( 46.8%) frame_23010.png
 → QR code detected in frame_23010.png: [redacted-qr-link]
[591/935] ( 63.2%) frame_22980.png
 → QR code detected in frame_22980.png: [redacted-qr-link]
[871/935] ( 93.2%) frame_23070.png
 → QR code detected in frame_23070.png: [redacted-qr-link]
[915/935] ( 97.9%) frame_22950.png
 → QR code detected in frame_22950.png: [redacted-qr-link]
[925/935] ( 98.9%) frame_23040.png
 → QR code detected in frame_23040.png: [redacted-qr-link]
[935/935] (100.0%) frame_3810.png

Summary of detected QR codes:
 - frame_23130.png: [redacted-qr-link]
 - frame_23100.png: [redacted-qr-link]
 - frame_22920.png: [redacted-qr-link]
 - frame_23010.png: [redacted-qr-link]
 - frame_22980.png: [redacted-qr-link]
 - frame_23070.png: [redacted-qr-link]
 - frame_22950.png: [redacted-qr-link]
 - frame_23040.png: [redacted-qr-link]

🎁 What Was the QR Code?

For the curious: the QR code led to a giveaway (for US residents) for a robot of your own, with an MSRP of $1,999. Not a bad prize for those with sharper eyes than mine!

😂 The “Aha!” (or “D’oh!”) Moment

After all that, I finally found the QR code… and wow, it was super obvious once I looked at the timestamp. I guess my superpower is missing things that are right in front of my face. Next time, maybe I’ll just pause the video and squint really hard. Or not. Thank you, Python, for saving me from myself!

🏁 Conclusion

With a little scripting, a lot of frames, and a healthy dose of self-mockery, I found the hidden QR code in Jerry Rig Everything’s video! If you want to try this yourself, all you need is yt-dlp, ffmpeg, Python, and OpenCV. Happy hunting—and may your eyes be sharper than mine!