Lead AI Engineer

Lead AI Engineer

Location: Pune/Mumbai/Hyderabad/Chennai/Bengaluru/Gurugram

AI Engineer responsible for designing, building, and deploying production-grade AI solutions on Multi Cloud especially Google Cloud Platform (GCP), with a focus on Document AI (DocAI), Agentic AI workflows, and Retrieval-Augmented Generation (RAG). The role will develop scalable pipelines and AI services using multiple AI offerings to enable semantic search, document understanding, and intelligent automation, partnering closely with product, engineering, data, and risk stakeholders to deliver secure, reliable outcomes.

- Design and implement end-to-end DocAI solutions (document ingestion, classification, extraction, validation, and human-in-the-loop review) for structured and unstructured documents.  
- Build and orchestrate Agentic AI systems (tool use, planning, memory, guardrails) to automate multi-step business processes and integrate with enterprise systems/APIs.  
- Develop RAG architectures for enterprise knowledge retrieval, including chunking strategies, embedding generation, indexing, reranking, and response grounding/citation.  
- Implement and operate vector database solutions including schema design, indexing, and performance tuning.  
- Develop scalable AI services based on business requirements from the Internal Audit.
- Establish evaluation frameworks for LLM/RAG (quality, hallucination/grounding, latency, cost), and implement monitoring/observability in production.  
- Apply security, privacy, and responsible AI practices (data handling, access controls, prompt safety, model governance) aligned to organisational standards.  
- Collaborate with architects and platform teams to define reference architectures, reusable components, and CI/CD pipelines for AI delivery.  
- Produce clear technical documentation, runbooks, and knowledge transfer; support incident triage and continuous improvement.  
- Contribute to engineering best practices: code reviews, testing, performance optimisation, and reliability engineering.

- Proven experience building and deploying AI solutions in production, including LLM-based applications.  
- Strong hands-on experience with DocAI/document understanding (OCR, extraction, layout parsing) and building document processing pipelines.  
- Solid experience implementing RAG systems end-to-end (retrieval, embeddings, vector indexing, reranking, grounding, citations).  
- Experience building Agentic AI workflows (function/tool calling, orchestration frameworks, state/memory management, guardrails).  
- Strong knowledge of vector databases and semantic search concepts; ability to tune for relevance, latency, and scale.  
- Proficiency in Python (must have); experience with common AI frameworks (e.g., LangChain/LlamaIndex or equivalent), and API/service development (FastAPI/ReactJS).  
- Strong GCP experience (Good to have) incl. GenAI capabilities, Cloud Run/GKE, BigQuery, Pub/Sub, Cloud Storage, IAM, VPC/networking basics, logging/monitoring.  
- MLOps/LLMOps experience: CI/CD, model/version management, automated testing, evaluation pipelines, and production monitoring.  
- Software engineering fundamentals: data structures, system design, secure coding, unit/integration testing, and performance optimisation.  
- Experience working in regulated environments and applying security/privacy controls (PII handling, encryption, access management) is preferred.  
- Bachelor’s/Master’s in Computer Science, Engineering, Data Science, or equivalent practical experience.