I'm a software engineer who ended up specializing in AI/RAG systems almost by necessity at InnovX (OCP Group),I was given the problem of making a 200GB+ SharePoint corpus actually queryable through a multi-agent RAG system, and that turned into the core of what I do now: LangGraph, pgvector, retrieval architecture, the whole pipeline from ingestion to evals .
What makes me different is that I don't stop at "it works." Test-Scale, a distributed load-testing platform I built in Go, exists because I wanted to actually measure system behavior under load with real numbers, not guesses — it found a generator-saturation bottleneck at 1.4k req/s and I fixed it, taking throughput to 1,889 req/s at 11ms p95 with zero errors. Citebench is the same instinct applied to RAG itself: an eval CI pipeline so retrieval regressions fail the build instead of shipping silently. I build the tool that proves the claim, not just the claim.
I'm looking for a fully remote Senior/Staff-track engineering role centered on AI/LLM systems or distributed backend architecture — somewhere I can own hard retrieval or infra problems end-to-end, not just implement someone else's spec. Remote is a firm requirement for me, not a preference