What You'll Do
● Independently design, build, deploy, and operate AI-powered features, APIs, services, and workflows.
● Develop production LLM applications using structured outputs, tool/function calling, RAG, agentic workflows, and context-management patterns.
● Design retrieval pipelines across structured and unstructured enterprise data, including chunking, embeddings, vector/hybrid search, ranking, and grounding strategies.
● Evaluate and integrate models from multiple providers based on quality, latency, reliability, security, and cost trade-offs.
● Build reusable components within your pod's codebase for prompts, tools, retrieval, and evaluations, and flag candidates for cross-pod standardization.
● Define and maintain automated evaluations and regression tests for AI quality, groundedness, task success, and failure modes.
● Implement production controls including observability, caching, retries, fallbacks, rate limits, human-in-the-loop flows, and cost controls.
● Troubleshoot production issues and improve reliability, latency, quality, and operating cost using measurable signals.
● Participate in technical design reviews and contribute to AI engineering standards and documentation.
● Support and mentor AI Engineer I teammates through reviews, pairing,



and knowledge sharing.

Technical Skills & Technology Stack
● Programming: Solid Python and software engineering fundamentals; working knowledge of Java or JavaScript/TypeScript is useful for platform integration.
● Data: Strong SQL; Snowflake or equivalent cloud data platforms; structured and unstructured data processing.
● Generative AI / LLMs: Hands-on experience with major model APIs, structured outputs, tool/function calling, model selection, and prompt/context design.
● AI Application Development: RAG, embeddings, vector/hybrid search, retrieval/ranking, agentic workflows, context management, and basic memory patterns.
● Frameworks: Practical experience with LangGraph, LangChain, LlamaIndex, or equivalent orchestration libraries; ability to work without framework lock-in.
● APIs & Integration: REST APIs, JSON, asynchronous/service integration patterns; familiarity with MCP and tool ecosystems is preferred.
● Cloud & Engineering: AWS or equivalent cloud, Docker, Git, CI/CD, automated testing, logging, and production debugging.
● Production AI: Evaluations, observability/tracing, guardrails, caching, fallbacks, latency optimization, and token/cost management.

📌 AI Engineer II (Ahmedabad)
🏢 Rekruton Technologies
📍 Ahmedabad