Description

Design and operate scalable LLM orchestration, RAG pipelines, and vector database infrastructure for production AI platforms.

This role is on-site.

Responsibilities

• Operationalize LLM workloads including prompt management, model routing, A/B testing, and safety guardrails.
• Build end-to-end RAG systems covering ingestion, chunking, embeddings, retrieval, and generation with hybrid search optimization.
• Design and maintain vector databases (Pinecone, FAISS, Milvus, or Weaviate), tuning indexing strategies for recall and latency.
• Develop scalable microservices and REST APIs using Python, FastAPI, or Flask, deployed via Docker and Kubernetes.
• Implement CI/CD pipelines, observability tooling, and data privacy controls including PII masking and audit logging.

Required Skills

• 7+ years of experience in software engineering with advanced Python skills (async, SOLID principles).
• Strong expertise in LLM Ops: orchestration frameworks (LangChain, LangGraph), evaluation metrics, and monitoring.
• Practical experience building production-grade RAG pipelines with hybrid retrieval (BM25 + dense search) and reranking.
• Hands-on experience with at least one vector database platform and understanding of ANN configurations.
• Experience with Docker, Kubernetes, and CI/CD tools (Jenkins, GitHub Actions, or Azure DevOps).
• Ability to design secure, scalable APIs with OpenAPI/Swagger standards, rate limiting, and structured error handling.

Preferred Skills

• Familiarity with model ecosystems such as OpenAI, Azure OpenAI, Anthropic, Cohere, Llama, or Mistral.
• Understanding of cost optimization strategies including caching (Redis) and token usage monitoring.

Key Skills

Education

Any Graduate