EXL—Data Scientist

Job description

Description: Key Responsibilities Technical Execution • Build and maintain high-performance REST/WebSocket APIs using FastAPI (Pydantic v2) • Implement and optimize agentic AI systems using frameworks like LangGraph, Deep Agents, AutoGen, and LangChain • Architect real-time, event-driven microservices using messaging queues like Apache Kafka • Design clean, testable, maintainable services using SOLID principles, Python async, and type hints • Integrate and optimize SQL, NoSQL, and vector databases (Postgres, MongoDB, ChromaDB, Pinecone) • Run LangGraph/Deep Agents workflows in production with checkpointing, persistence, and human-in-the-loop controls, backed by LLM observability and evaluation tooling (e.g., LangSmith, Langfuse) • Apply LLM safety guardrails (prompt-injection mitigation, PII handling, content moderation) in line with Banking, Insurance, and Healthcare compliance requirements • Stay current with emerging trends in GenAI, deep learning, and AI orchestration frameworks Requirements: Minimum Qualifications • Bachelor's degree in Computer Science, Data Science, or related field • 3+ years of total professional experience, including: • 3+ years of hands-on Software Engineering experience in Python, FastAPI and relevant tech stack • 2+ years working specifically with GenAI and LLMs (GPT, Claude, LLaMA, etc.) • Track record of shipping production ML/AI products, with strong prompt-engineering skills, systems-level thinking, and the ability to diagnose and resolve production failures • Strong software engineering background with expertise in OOP and SOLID principles • Proficiency in Python 3.11+ (async/await, type hints, modern Python patterns, strict type checking) • Experience with agentic frameworks (LangChain, LangGraph, AutoGen, or similar) • Proven track record building production REST/WebSocket APIs and microservices • Experience with message streaming platforms (Kafka, Pulsar, or similar) • Strong knowledge of databases: SQL, NoSQL, and vector databases • Working knowledge of RAG pipelines (embeddings, chunking, retrieval) and LLM observability/evaluation tools (LangSmith, Langfuse, or similar) Responsibilities: Key Responsibilities Technical Execution • Build and maintain high-performance REST/WebSocket APIs using FastAPI (Pydantic v2) • Implement and optimize agentic AI systems using frameworks like LangGraph, Deep Agents, AutoGen, and LangChain • Architect real-time, event-driven microservices using messaging queues like Apache Kafka • Design clean, testable, maintainable services using SOLID principles, Python async, and type hints • Integrate and optimize SQL, NoSQL, and vector databases (Postgres, MongoDB, ChromaDB, Pinecone) • Run LangGraph/Deep Agents workflows in production with checkpointing, persistence, and human-in-the-loop controls, backed by LLM observability and evaluation tooling (e.g., LangSmith, Langfuse) • Apply LLM safety guardrails (prompt-injection mitigation, PII handling, content moderation) in line with Banking, Insurance, and Healthcare compliance requirements • Stay current with emerging trends in GenAI, deep learning, and AI orchestration frameworks