AI engineer shipping production LLM systems into regulated, air-gapped customer environments.

Deployments in enterprise production systems (automated QA compliance, conversational analytics, real-time agent assist).
Open-source tools spanning evaluation, agentic compliance, RL environments, and small-model fine-tuning.

New: dental-desk: 11-state FSM voice agent with six guardrail layers. 1,500 lines of TypeScript, no SDK abstractions. Tool scoping makes hallucination structurally impossible. Live demo · GitHub.

Latest write-up: Building MCP servers that survive a regulator's audit: 7-stage compliance pipeline, 15-case golden-set eval, 100% pass.

For companies shipping LLM features into regulated workflows: Automated compliance and audit systems, LLM evaluation harnesses, and automated QA pipelines for collections, direct sales, BPO, and adjacent verticals. Book a 20-min call

For teams hiring forward-deployed or founding AI engineers: On-premise inference, ambiguous customer scoping, regulated verticals, full-stack ownership. Open to remote and relocation. ree2raz@proton.me