Phil SpringerIndependent AI Engineer

Independent AI Engineer / Phoenix / since 2010

Multi-agent AI systems that run real operations.

I orchestrate frontier and self-hosted models into agent fleets that ship code, run infrastructure, and execute operations. A different-lineage model adversarially reviews every output before it ships. Nothing goes out on one model's say-so.

who

Sixteen years, my own shop.

Computer scientist by training, B.S. Computer Science with a Business Administration minor from Southern Oregon University, 2010, and an independent technology founder ever since. I have run my own S-corporation for sixteen years, and over the last few I rebuilt the entire operation around applied large language models.

I own the whole stack and ship end to end. Everything I run is hardened against my own money, not a client deck.

method

Different lineage, different blind spots.

A model checking its own work has the same blind spots as the work. Ask it to review itself and it agrees with itself, confidently, every time. So the reviewer is always a model from a different lineage: different training, different failure modes.

When the two disagree, I have found the hole. When they agree, I have a real second opinion instead of an echo. I treat every model output like production code: measured against a locked baseline, assumed wrong until an independent check proves it right.

stack

What I actually run.

Orchestration
Multi-agent systems coordinating frontier models (Claude, GPT/Codex) with self-hosted open-weight LLMs over MCP. Agentic coding is the daily driver, not an experiment.
Local inference
Deployment and fine-tuning on Hermes, my 512 GB Apple-silicon node. Real compute I own and run, not a rented endpoint.
Languages
Polyglot and stack-agnostic. Python, PHP/Laravel, JavaScript and TypeScript, Next.js, React, Node, SQL, shell, on a CS foundation reaching Java, C, C++, and Go. I ship in whatever the problem needs.
Infrastructure
A self-hosted fleet of Linux servers on Contabo plus a rebuilt Intel i9 running my private Git (Forgejo) and automation. Tailscale, Caddy and nginx, Docker, Cloudflare, MySQL/MariaDB. My own Git, my own servers, my own models.
Model work
Prompt and context engineering, evaluation, adversarial red-teaming, RAG, and technical and code data generation.

open to

AI engineering, model evaluation, applied LLM work.

Remote, or Phoenix hybrid. You get someone who already lives inside these systems at production stakes every day, not someone ramping into them. I build in public, and the work speaks before I do.