Projects are for proving something. Each entry starts with a brief: the problem, the hypothesis, the goals, the outcome, the tools used, and the notes that matter after the work is done.
tooling
Foundation
Semantic Drift Watcher
A monitoring layer that compares live retrieval output against stored intent baselines so drift shows up before the model starts sounding wrong.
tooling
Foundation
Cost-Aware Prompt Router
A FastAPI middleware layer that inspects task shape and routes requests to the lowest-cost model that can still do the job well.
experiment
Foundation
Patch-and-Retry Sandbox
A contained patch-and-retry loop where an agent observes its own failures, proposes changes, and keeps iterating until the test passes or the budget runs out.
prototype
Frontier
BM25 + Dense + Reranking
A retrieval pipeline that blends BM25, dense vectors, and a cross-encoder reranker so exact-match precision and semantic recall can work together instead of competing.
prototype
Frontier
Outbound PII Filter
A proxy layer that runs a regex pass for structured PII and a BERT-NER pass for unstructured entities, combining both before any outbound model call.
live system
Live System
Public Runtime + Recap Pipeline
A public voice-agent project with signed access, recap lookup, transcript handling, rate limiting, and a private regression harness behind the scenes.
experiment
Local AI
Local Gemma 4 + LoRA on Apple Silicon
A practical test of Gemma 4 on a 16GB MacBook M2: run the small edge model locally with Ollama and Metal, then shape a LoRA tuning path for private document and audio extraction.