Raspberry Pi + Web Developer for AI Kiosk (6-Hour On-Site Sprint, Coconut Grove)
Role details
Job location
Tech stack
Job description
Set up and configure the Raspberry Pi 5 (Raspberry Pi OS, Firefox ESR kiosk mode, autostart on boot, display rotation to portrait). Get the USB webcam negotiating proper resolution on the Pi at the driver level (v4l2 / MJPEG configuration). Confirm browser-based face detection runs reliably on the Pi's hardware. Run the full flow end to end on the device: face detection, photo capture, live API call, on-screen results, QR handoff. Work from an existing setup script and validation checklist (both already written). Document what you did and flag anything that needs a follow-up.
Requirements
Do you have experience in Web development?, Hands-on Raspberry Pi experience (provisioning, Linux command line, SSH, kiosk-mode setup). Comfort with the Linux camera/video stack (v4l2-ctl, USB webcam configuration). Working knowledge of web technologies (HTML/CSS/JavaScript) enough to read and debug an existing single-page app. Browser dev tools and console debugging. Git/GitHub.
Nice to have
Experience with kiosk or digital-signage deployments. Familiarity with MediaPipe or browser-based computer vision. Experience integrating web apps with REST APIs.
This could turn into ongoing work
I run several AI and Raspberry Pi projects out of this location. If this sprint goes well, there's steady project work available for the right person. I'm looking for someone reliable, sharp, and able to work independently, not just for one day but as a go-to for hardware and AI builds.
Benefits & conditions
This is on-site in Coconut Grove, Miami FL (private home office). Local Miami candidates only. This is a contract, one-day project. Compensation is a flat $600 for the 6-hour on-site sprint, with potential for ongoing project work.
Pay: $600.00 per day
Application Question(s):
- Briefly describe the last time you got a USB camera working on a Raspberry Pi. What formats/resolutions did it report, what snag did you hit, and how did you solve it at the driver level (e.g. v4l2, MJPEG vs raw YUYV)?
- This is a finishing job on an already-built single-page app. How comfortable are you working only on the hardware/driver layer and debugging existing code without rewriting it? Give a brief example.
- This role is 100% on-site in a private home office in Coconut Grove (local Miami candidates only) for a single 6-hour block. Are you local and able to work on-site, and what is your availability over the next two weeks?
- You're an hour into a hardware problem on the Pi with no progress. Walk me through how you'd troubleshoot it and when you'd decide to flag it rather than keep trying things.