Role focus:

- Design Python pipelines for document ingestion, chunking, retrieval and generation.

- Extract structured information from complex PDFs, scanned records, tables and forms, with citations back to the source.

- Build validation, guardrails, evaluation and human-review workflows to improve accuracy.

- Develop model-independent components, agentic workflows and speech-to-text pipelines where needed.

- Work with the AI lead and product team; mentor engineers and support production reliability.

What we are looking for:

- 5+ years of software engineering experience with strong Python and production backend services such as FastAPI.

- Production LLM applications, tool calling and structured outputs with Pydantic.

- Production vector-database experience, preferably Qdrant; advanced RAG, hybrid retrieval and re-ranking.

- Document vision/layout models, evaluation datasets and measurable quality testing.

- Experience leading or mentoring engineers.

- Preferred: LoRA/PEFT, open-weight model serving, AWS, asynchronous processing, containers and regulated document workflows.

Contact:

MIYO Global, +91 83102 89665