Bosch Group is looking for an AI Engineer to join our team. You will build and operate LLM-powered solutions that transform legal, regulatory, and compliance document workflows. Your focus will be on designing and implementing GenAI and Agentic AI applications for complex document understanding, reasoning, and decision support.
What You'll Do
- Design and implement GenAI and Agentic AI applications for complex document understanding, reasoning, and decision support.
- Build and optimize RAG (Retrieval-Augmented Generation) pipelines tailored to regulatory and legal documents with high precision and grounded responses.
- Develop robust document ingestion and retrieval strategies including contextual chunking, embeddings, metadata enrichment, and semantic indexing.
- Implement reference identification, citation tracking, and traceability mechanisms for document-centric AI workflows.
- Optimize retrieval ranking, semantic search, and grounding to improve answer accuracy and reduce hallucinations.
- Integrate Knowledge Graphs (RDF/SPARQL) with LLM workflows for structured and unstructured reasoning.
- Orchestrate multi-step AI workflows using LangChain, LangGraph, or similar agent frameworks.
- Establish AI quality assurance and evaluation practices including retrieval evaluation, hallucination detection, LLM judge frameworks, and RAGAS-style scoring.
- Build, train, and fine-tune specialized NER and document understanding models.
- Ensure explainability, auditability, and compliance of AI outputs in regulated environments.
- Support end-to-end model lifecycle activities including experimentation, versioning, deployment readiness, and monitoring handover.
What We're Looking For
- 4–8 years of experience as an AI Engineer focused on building and operating LLM-powered solutions for legal, regulatory, and compliance document workflows.
- A strong emphasis on reference identification, citation grounding, retrieval quality, traceability, explainability, and evaluation in document-centric AI systems.
- A BE/B.Tech or Equivalent Degree.
- Strong hands-on expertise in Python (or Java), NLP, RegEx, SpaCy, NLTK, and transformer-based models.
- Proven experience with the following: Python (primary), Docker / Docker Compose, NLP / NLU, GenAI / LLM application development, Agentic AI, RAG (Retrieval-Augmented Generation), Embeddings, Contextual chunking strategies, Knowledge Graphs (RDF), SPARQL, Model lifecycle / ML application lifecycle, LangChain, LangGraph, Git, and AI QA / evaluation (e.g., RAGAS, LLM judges, retrieval and answer quality validation).
Nice to Have
- Experience with Java, Kubernetes, Dev Containers, or GitOps.
- Strong Documentation practices.
- Experience with Deepagents or similar advanced agent frameworks.
- Direct experience working with legal, regulatory, compliance, and policy documents.
- Understanding of requirements around auditability, explainability, and risk controls in AI systems.
