Start searching
Back to Jobs
Recently postedTechnology

Lead Generative AI Engineer/AI Architect

Vish Consulting Services, Inc.United States🇺🇸United States

Quick overview

SeniorityMid Senior
Work modeHybrid
LocationUnited States
FastAPIMongoDBGoogle CloudLLMPythonGenerative AIRESTAzureReactPostgreSQLTypeScriptLangChainLangGraphRAG

Job description

  • 12-13+ years of total IT experience. Should have deep knowledge in the IT domain and hands on experiece in software development, AI/ML, data science RAG pipeline.
  • 8+ years of hands-on AI/ML experience. The client clarified that the eight years must be in AI/ML. General software development experience does not count toward this requirement. Look for candidates who have personally built and shipped production AI applications.
  • Agent development and cloud. Screen for hands-on experience with Google Cloud Platform and/or Azure and ask which tools they used in production, particularly Google ADK, AutoGen, OpenAI models, LangChain, LangGraph, and LangSmith.
  • Production RAG pipelines. Candidates should be able to walk through an end-to-end pipeline they built, including how they used knowledge graphs and implemented hybrid and vector search. The team primarily uses MongoDB, and PostgreSQL is also acceptable. Expect a hands-on RAG exercise in the client interview.
  • Backend orchestration and custom MCP servers. Prioritize Temporal experience and ask candidates to describe a workflow they personally built using it. Validate hands-on Python, FastAPI, REST APIs, and event-driven design. Candidates must be able to explain how they built a custom Model Context Protocol (MCP) server and connected it to an agent.
  • Frontend development. Confirm 3+ years of React and TypeScript experience, including building modern chatbot interfaces and web components for prompt-driven interactions.
  • Testing and RAG/LLM evaluation. This is a must-have and one of the biggest gaps the client has seen. Ask what automated tests and quality controls the candidate personally put in place, which evaluation tools they used (RAGAS, DeepEval, Promptfoo, LangSmith, or similar), and how they measured quality and caught regressions before release. Expect the client to probe this in the interview.

Similar jobs

Apply