Senior Manager - RAG/LLM Specialist

- Design and manage multi-stage RAG pipelines delivering low latency and high retrieval relevance. - Drive fine-tuning initiatives (PEFT/LoRA) on open-source models for domain-specific performance. - Build automated evaluation frameworks (RAGAS) to measure accuracy, context precision, and recall. - Architect metadata filtering and hybrid search strategies in vector databases (Pinecone, Milvus). - Write Python pipelines for document parsing, text chunking, and embedding generation. - Systematically test and version-control prompts for reliable, repeatable model behaviour. What Makes You a Fit: - 3-7 years in ML/AI, software, or data engineering, including 1+ years hands-on with LLMs. - Strong Python with PyTorch/TensorFlow, LangChain, LlamaIndex, and embedding models. - Working knowledge of advanced RAG patterns (HyDE, parent-document retrieval) and prompt optimization. - Hands-on experience with LLM APIs (OpenAI, Anthropic). - Notice period: 30 days or less only. - Education: Bachelor's/Master's in Computer Science, Data Science, or related field. Why This Role: You own the retrieval layer of enterprise AI - the difference between a demo and a system businesses run on. Level (Senior Analyst / Senior Manager) is mapped to your experience and depth. .