We are hiring an AI Engineer (Generative AI) to build and productionize GenAI capabilities across product use cases. This role will focus on designing and developing LLM applications (RAG, agents/workflows, prompt orchestration), building evaluation and safety mechanisms, and integrating GenAI features into real product experiences. The engineer will partner with product, architecture, data engineering, and platform teams to ship secure, reliable, and cost-effective GenAI systems.

Key Responsibilities :

- Design and implement agentic GenAI systems : tool/function calling, plannerexecutor workflows, multi-step task decomposition, stateful sessions, and human-in-the-loop checkpoints.

- Build and maintain robust RAG systems : document processing, chunking, embeddings, vector indexing, hybrid retrieval, reranking, grounding and citation patterns.

- Implement orchestration and workflow engines for GenAI : routing, memory patterns, context assembly, retries, fallbacks, and safe degradation.

- Build tool integrations for agents : APIs to enterprise systems, structured action schemas, permissioning, audit logs, and sandboxing where required.

- Create evaluation harnesses for both RAG and agents : golden datasets, scenario-based tests, automated regression suites, and offline/online quality measurement.

- Implement safety and governance controls : prompt injection/jailbreak mitigation, PII handling/redaction, content filtering, secure logging, and policy-based controls.

- Optimize latency and cost : model selection, batching/caching strategies, prompt efficiency, token monitoring, and throughput tuning.

- Collaborate with backend/frontend teams to integrate GenAI services into products : APIs, authN/authZ, user context, feedback capture, and UX patterns for trust.

- Work with data engineering to curate and govern knowledge sources and ensure access controls and lineage for retrieval.

- Productionize GenAI services with platform teams : CI/CD hooks, monitoring/observability, incident response readiness, and operational runbooks.

- Document designs and contribute reusable components (agent templates, tool adapters, prompt libraries, evaluation suites).

Must have Skill Sets :

- Strong programming skills in Python, with solid software engineering fundamentals (APIs, testing, versioning, logging, error handling, performance basics).

- Practical experience building LLM applications using orchestration frameworks/patterns (e.g., LangChain/LlamaIndex/Semantic Kernel or equivalent approaches).

- Hands-on experience with agentic patterns :

a. tool/function calling and structured action execution

b. multi-step workflows, planning, and state management

c. reliability techniques (timeouts, retries, deterministic constraints, guardrails)

- Strong understanding of RAG fundamentals and tradeoffs : chunking, embeddings, retrieval tuning, reranking, grounding, citation strategies.

- Experience integrating LLM systems with enterprise backends : authentication/authorization, role-based access, audit logs, safe action execution.

- Working knowledge of LLM evaluation : scenario tests, regression testing, prompt/model comparisons, and feedback-driven iteration.

- Working knowledge of security/privacy for GenAI : prompt injection risks, PII handling, safe logging, and data usage boundaries.

- Clear communication and documentation skills; ability to translate ambiguous requirements into production-ready systems.

Good to have Skill Sets :

- Experience with vector databases/search stacks (pgvector, FAISS, Elastic/OpenSearch, Pinecone, Milvus) and hybrid retrieval.

- Experience implementing routing and mixture-of-experts patterns (model/prompt routing by intent, confidence, or cost).

- Familiarity with workflow engines (Temporal, Airflow/Prefect for orchestration adjacent needs) and event-driven patterns.

- Exposure to fine-tuning techniques (LoRA/PEFT), synthetic data generation, or domain adaptation strategies.

- Familiarity with observability for LLM apps : retrieval quality metrics, tool-call success rates, token/cost, latency, and error budgets.

- Experience deploying AI services with containers and CI/CD in cloud/on-prem environments (in partnership with platform teams).

Qualifications :

- Bachelors/Masters in Computer Science, Engineering, AI/ML, or related field (or equivalent practical experience).

- 3 to 7 years of overall experience, with 1+ year delivering GenAI/LLM features (RAG and/or agentic workflows) in production or production-like environments.

- Demonstrable work artifacts (internal products, prototypes, or repositories) showing agent/tool integrations, RAG pipelines, evaluation setup, and production integration.

Job Location : Mumbai

Whats in store for you?

- Be an equal parent

- Childcare benefits for the birthing parent, commissioning parent (in case of surrogacy) or adoptive parent, and their partners

- 6 months of paid leave for primary caregivers, flexible work options on return for primary caregivers

- 2 months paid leave for secondary caregivers

- Caregiver travel for primary caregivers to bring a caregiver and children under a year old, on work travel

- Coverage for childbirth and fertility treatment

- No place for discrimination at Godrej

- Gender-neutral anti-harassment policy

- Same sex partner benefits at par with married spouses

- Coverage for gender reassignment surgery and hormone replacement therapy