About

I am Quinn Favo. I build AI systems that hold up in production.

A software engineer focused on building pragmatic, high-performance applications and robust AI systems.

My background spans across cloud infrastructure, DevOps, and mobile development. I specialize in bridging the gap between data science and platform engineering, helping enterprises deploy secure, scalable AI models on their own infrastructure.

I believe in writing clean, maintainable code and avoiding the hype cycle. Whether I'm orchestrating a Kubernetes cluster for an LLM deployment or optimizing a native Kotlin app for Wear OS, my focus is always on performance, security, and reliability.

I am based in the New York area and work remotely with US-based clients.

Engineering philosophy

Three rules I don't bend.

Performance is a feature.

Whether it's a constrained edge device or a GPU cluster, efficient code respects resources.

Own your infrastructure.

In the age of AI, data privacy, compliance, and on-premise deployments are critical for enterprises.

Build in public.

I share my learnings, architectures, and code via YouTube and technical articles to help the community.

What I work on

Infrastructure, agents, automation, and apps.

AI infrastructure

Private and hybrid LLM deployments, from a single GPU box to a Kubernetes cluster.

Agent systems

Agent teams with owners, budgets, eval gates and human approval, built to be handed over.

DevOps and automation

CI/CD, observability, and the scripts that keep systems boring in the best way.

Android and Wear OS

Native Kotlin apps I design, build and ship myself on Google Play.

How I ship my own apps

The same crew pattern I build for clients runs my app studio.

Feedback becomes owned issues, tests can send work back, and nothing reaches Google Play until I have reviewed it.

Goal

Turn user feedback into the next app release

Crew

Audit log

  1. 09:10:58handoff docked
  2. 09:10:58ops docked to recharge
  3. 09:10:58eval back on the gate
  4. 09:10:11ops built the release and sent it to Google Play
  5. 09:09:24ops woke on heartbeat
Illustrative. Simplified from my real release flow.
  1. a 4-star review asked for a new setting
  2. spec opened #41, #42, #43 with one owner each
  3. retriever checked out #41
  4. eval loaded the test suite
  5. sentinel loaded the policy rules
  6. scribe woke on heartbeat
  7. retriever found the settings screen and its tests for #41
  8. retriever brought the settings code to the scribe
  9. handoff rolled over to carry the draft
  10. handoff built the setting for #41
  11. handoff opened a PR for #41
  12. retriever checked out #42
  13. eval: #41 broke a UI test on small screens
  14. sentinel: no new permissions requested
  15. retriever found the tile code for #42
  16. handoff carried #41 back to the scribe
  17. handoff fixed the small-screen layout
  18. handoff updated the PR for #41
  19. router checked progress on #42
  20. eval: #41 passes unit, UI and accessibility checks
  21. sentinel cleared the fix
  22. Quinn reviewed and approved #41
  23. ops woke on heartbeat
  24. ops built the release and sent it to Google Play
  25. eval back on the gate
  26. ops docked to recharge
  27. handoff docked

Build in public

Watch how it gets built, or hire me to build yours.

I publish the architectures, tools and mistakes on YouTube and in technical articles. If you want the same thinking applied to your systems, start with a free fit call.