Principal AI Engineer

BuzzBoardAI products

Source API remote eligibility restrictions: IndiaFull-TimePrincipal

Salary not disclosed

Apply NowOpens the employer's application page

Job Details

Experience: 5+ years in engineering, AI, ML, or data-product work; 3+ years hands-on with GenAI / LLM systems
Required Skills: DockerPythonRESTful APIsGenerative AI

Have 5+ years in engineering, AI, ML, or data-product work, including 3+ years hands-on with GenAI/LLM systems.
Have production GenAI experience building or scaling systems that carried real traffic.
Know LLMs and SLMs, prompt engineering, structured outputs, and tool and function calling.
Have hands-on experience with at least two major LLM ecosystems and be able to reason about their tradeoffs.
Have hands-on RAG experience with vector stores such as Chroma, Pinecone, Weaviate, or FAISS, embeddings, semantic search, and measured retrieval quality.
Have hands-on experience with an agentic framework such as LangGraph, CrewAI, AutoGen, or Semantic Kernel, and understand memory, state, tool integration, and failure handling.
Have strong Python skills and be comfortable with REST APIs, Docker, a cloud platform, and basic CI/CD.
Have demonstrated evaluation experience, including regression testing, hallucination checks, schema validation, or output scoring.
Have a track record guiding a small team or owning AI architecture end to end.
Be comfortable creating structure in a fast-moving environment with shifting requirements.

Design shared AI system architecture and reusable patterns for content generation, business intelligence, recommendations, and agentic workflows.
Design agents that reason, call tools and APIs, manage state, checkpoint, and hand tasks off cleanly.
Build guardrails that keep agent workflows grounded, predictable, and able to fail safely.
Design end-to-end RAG pipelines, including chunking, embeddings, metadata, reranking, and retrieval evaluation.
Own model selection and define fallback and model-switching behavior when a provider degrades.
Build evaluation frameworks and regression testing for output quality, hallucination rate, schema adherence, latency, failure rate, and inference cost.
Partner with engineering and platform teams to make systems deployable, observable, and maintainable.
Package services with Python, FastAPI/Flask, and Docker when appropriate, and diagnose production AI failure modes.
Mentor GenAI engineers, review designs and workflows, and translate product requirements into architecture with measurable acceptance criteria.

View Full Description & ApplyYou'll be redirected to the employer's site