Boston, MA · open to full-time
ML Platform Engineer in Boston
Abhishek Sagar Sanda is an ML platform engineer who builds the rails under production AI - retrieval corpora, voice agent runtimes, evaluation loops, and failover - so models become dependable services.
What an ML platform engineer owns
An ML platform engineer turns model experiments into operable systems: ingestion and indexing for retrieval, orchestration for agent loops, latency budgets, observability, and the path from staging to a live call or product surface.
As an ML platform engineer, Abhishek focuses on platforms that keep voice and RAG workloads honest - ChromaDB-style corpora with citations, LiveKit and Twilio runtimes near 800-millisecond response windows, and fences that fail closed when a model or network path breaks.
Platform proof from shipped systems
- Northeastern University Assistant - RAG platform over 80,000+ pages with citations, the retrieval half of grounded answers.
- AutoAce (YC F25) - voice-agent OS for dealerships: latency cut 8.5s → 3.9s, backfilled sessions, three-layer LiveKit fence.
- BitVoice Pay - voice-first Lightning wallet on a live dial-in; agent runtime on Twilio end to end.
- AI Interview Coach IVR - multi-turn interview coaching on a live call path.
Hire an ML platform engineer
Based in Boston and available for full-time roles, Abhishek joins teams that need an ML platform engineer to take voice and retrieval workloads from pilot to operable service. Related role pages: AI agent engineer, full-stack AI engineer, forward deployed engineer. See selected work, the resume, or book a call.