AI / ML Engineer

Role overview This mid-to-senior engineering role focuses on building the AI reasoning layer behind prescription intelligence for Indian clinical practice. The work involves designing and shipping production models and pipelines that read prescriptions, interpret clinical context, and surface decisions clinicians can trust, with daily clinical review of outputs.

Responsibilities - Design and build data pipelines covering OCR, entity extraction, and normalisation of prescription content - Fine-tune and evaluate large language models on clinical reasoning tasks using rigorous offline metrics - Own model serving in production, including latency budgets and per-inference cost optimisation - Build evaluation harnesses in collaboration with clinical advisors to prevent regressions and ensure safety - Maintain reliability and performance of inference infrastructure end-to-end

Requirements - Three or more years building machine learning systems that run in production - Strong proficiency in Python and a modern ML stack such as PyTorch or HuggingFace, or equivalents - Hands-on experience with LLM fine-tuning, retrieval methods, or structured generation - Comfort deploying models behind low-latency APIs at meaningful scale - Sound engineering judgement around evaluation, monitoring, and iteration

Nice to have - Prior experience with medical NLP or biomedical datasets - Familiarity with the Indian pharmacopeia or ICD-10 coding systems - Published research, open-source contributions, or a verifiable Kaggle track record

Benefits and work setup - Full-time, remote-first position based in India - Target stack includes Python, PyTorch, HuggingFace, Postgres, and Ray - Structured hiring process covering application review, an introductory call, a craft interview, and a final conversation, with offers typically issued within about a week of the last stage