2 days ago
Bengaluru, IndiaStaff+
Responsibilities
- Build Python backend services with FastAPI and Pydantic for retrieval, extraction, and agent features.
- Create data ingestion pipelines for CSV, JSON, Excel, and PDF files, including cleaning, deduplication, database loading, vector storage, and data-quality checks.
- Implement RAG pipelines with chunking, embeddings, vector search, citations, and appropriate unknown-context responses.
- Develop structured extraction, tool-calling agents, human approval workflows, React interfaces, dashboards, and Streamlit or Gradio prototypes.
- Integrate with client systems through APIs, webhooks, scheduled file exchanges, PostgreSQL, SQL Server, and Oracle.
- Build evaluation datasets, measure precision, recall, F1, retrieval, and answer-quality metrics, and integrate automated tests into CI.
- Containerize and deploy services with Docker, add logging, metrics, and traces, and prepare runbooks and handoff documentation.
- Collaborate directly with clients through working sessions, feature demonstrations, feedback capture, and technical field notes.
Requirements
- Bachelor’s degree in computer science, engineering, mathematics, statistics, or a related field, or equivalent demonstrated engineering experience.
- Practical Python 3.10+ development with type hints, error handling, classes, functions, and Pydantic models.
- Experience writing pytest tests, using Git branches and pull requests, building FastAPI endpoints, and calling APIs with timeouts and retries.
- HTML, CSS, JavaScript, and React fundamentals, including components, props, state, hooks, API calls, and loading and error handling.
- SQL knowledge including joins, GROUP BY, CTEs, primary keys, foreign keys, and simple schema design.
- Experience cleaning CSV or JSON data with pandas and deploying at least one project for other users.
- Dockerfile creation and container deployment experience.
- Understanding of LLM tokens, context windows, cost, structured prompting, and at least one RAG pipeline with citations.
- Knowledge of train, validation, and test splits, data leakage, precision, recall, F1, confusion matrices, and imbalanced-data accuracy limitations.
- Ability to explain technical trade-offs clearly in written and spoken English.
- Preferred qualifications include a public deployed RAG or agent portfolio, internships, hackathons, open-source contributions, domain exposure to finance, insurance, healthcare, or supply-chain documents, Java or C# familiarity, Figma exposure, knowledge graphs, or vision-language models.
Benefits
- Hybrid work arrangement in Bengaluru with a day shift in India.
- Full-time regular employment.
- Training and mentoring in advanced full-stack AI engineering, cloud operations, enterprise integration, retrieval, agents, evaluation, and security.
- Hands-on experience solving enterprise client problems and opportunities for career development.
- Values-driven, inclusive work environment with equal-opportunity employment practices.
Tech Stack
Categories
Forward Deployed
About Genpact
Genpact is a global professional services and technology firm that runs and transforms mission-critical operations for large enterprises through consulting, managed services, and AI- and analytics-driven platforms. It serves industries such as banking, insurance, healthcare, and manufacturing/supply chain, modernizing finance, risk, customer service, and back-office processes. Headquartered in New York and listed on the NYSE (G), Genpact originated within GE before becoming an independent public company.