About this role
Job Summary We are looking for an AI Engineer to build, evaluate, and ship GenAI-powered applications in production. You will turn foundation models into reliable, observable features - building Python APIs, RAG pipelines, and agentic workflows, proving quality with evals, and deploying with strong reliability and cost discipline. You will work across development, experimentation, and deployment alongside ML, product, and platform teams. Job Requirements Build / Experiment Design Python APIs (FastAPI/Flask) and integrate LLMs, embeddings, RAG, and agents (LangGraph, Google ADK, LangChain/LlamaIndex) into production. Implement tool/function calling and MCP; apply prompt and context engineering. Build data ingestion, preprocessing, and inference pipelines with clear observability. Own the eval harness: write evals before shipping (RAGAS, DeepEval, or LangSmith) covering faithfulness, retrieval quality, latency, and cost. Run prompt/model A/B tests and tune retrieval (chunking, hybrid search, re-ranking). Explore light fine-tuning (LoRA/QLoRA) when it beats prompting or RAG. Ship & Operate (Good to have) Deploy on Docker/Kubernetes with CI/CD, versioning, and safe rollout. Add logging, metrics, and tracing (Grafana, OpenTelemetry) and optimize latency, throughput, and cost (caching, batching, streaming). Write tests, review code, and support production incidents. What We're Looking For (Required) 2-3 years of software development experience, including ML/AI engineering. Strong Python and production API experience (FastAPI, Flask, or Django REST). Hands-on GenAI app experience: LLM integration (OpenAI, Anthropic, Azure OpenAI, Bedrock, or Vertex), RAG, and at least one agent framework with tool calling. Prompt/context engineering and an evaluation mindset (RAGAS/DeepEval/LangSmith). Experience with a vector database (pgvector, Pinecone, Weaviate, Qdrant, or FAISS). Awareness of guardrails, prompt-injection defense, PII handling, and responsible AI. Async programming; SQL/NoSQL, caching (Redis), and message queues (Kafka, SQS, or similar). Docker/Kubernetes, CI/CD, and Git. Education Education and General Skills: Bachelor’s degree in engineering/ statistics/ data science (or related). Master’s degree is a plus 4-8 years of professional experience in Software Development / ML Engineering