Onsite: Gurgaon | 6 Days Working
Experience: 2+ years
Significant: We’re specifically looking for someone who has spent their experience building and shipping real AI agents, agentic systems, or RAG systems used by real users at scale.This is not a general software engineering role. If your experience is primarily in backend/software engineering and you’ve only recently started working with LLMs or AI, please don’t apply.
We’re looking for someone whose ~2+ years of experience is predominantly hands-on AI engineering - building production agents, RAG pipelines, LLM systems, orchestration, evaluation, and related infrastructure.
Tech Stack: Python, LLM APIs, RAG, Vector DBs, Model Training/Fine-tuning, Docker, Kubernetes, Redis, MLOps, FastAPI, Postgres, AWS
About the Role
Build and ship production LLM systems - not prototypes. You'll own the full lifecycle: retrieval architecture, agent/orchestration design, model training/fine-tuning, evaluation, and deployment for features with real users and real failure consequences.
Responsibilities
- Design and build RAG pipelines — chunking, embeddings, vector search, re-ranking
- Build agents and orchestration logic from scratch (no framework crutch) — tool-use, multi-step reasoning, state management
- Train and fine-tune models where off-the-shelf LLMs fall short — dataset curation, fine-tuning, LoRA, evaluation of trained models
- Set up eval harnesses and benchmarks to catch regressions before users do
- Implement guardrails, hallucination detection, and prompt-injection defenses
- Optimize for cost, latency, and context-window efficiency — caching, streaming, Redis
- Design backend APIs and data models independent of the AI layer
- Containerize and deploy services with Docker/Kubernetes; build MLOps pipelines for model versioning, monitoring, and rollout
- Run A/B tests and iterate on prompt/model performance
- Maintain observability and tracing across LLM pipelines
Required Skills
- 2+ years of hands-on experience
📌 Generative AI Engineer (Kochi)
🏢 Lawvek
📍 Kochi