Experience Level: 4–6 Years
Location: Bangalore, UB City office
Notice Period: Immediate
About the Role:
We are seeking a skilled and passionate RAG Prompt Engineer to join our AI/ML team. This role focuses on designing, developing, and optimizing prompts and pipelines for Retrieval-Augmented Generation (RAG) systems using Large Language Models (LLMs). The ideal candidate has a strong background in natural language processing, prompt engineering, and information retrieval, with hands-on experience building scalable LLM-powered applications.
Key Responsibilities:
- Design and implement prompt strategies for RAG-based systems using leading LLM frameworks (e.g., OpenAI, Hugging Face, Cohere).
- Integrate vector databases and retrieval systems (e.g., FAISS, Pinecone, Weaviate) with LLMs for accurate and context-aware responses.
- Fine-tune or instruct-tune LLMs (e.g., LLaMA, GPT-4, Mistral) for domain-specific applications.
- Optimize query performance and retrieval accuracy in vector search engines.
- Evaluate and iterate on prompts using metrics such as relevance, coherence, factual accuracy, and latency.
- Collaborate with product and research teams to deploy RAG pipelines in production environments.
- Stay up to date with the latest advancements in LLMs, retrieval methods, and generative AI.
Required Skills & Qualifications:
- 4–6 years of experience in NLP, ML, or AI-focused roles, with at least 1–2 years in prompt engineering or LLM application development.
- Proven experience with RAG architectures and implementation.
- Proficiency in Python and experience with libraries like LangChain, LlamaIndex, or similar orchestration tools.
- Experience with LLM APIs (OpenAI, Anthropic, Cohere, etc.) and open-source LLMs (Mistral, LLaMA, Falcon, etc.).
- Strong understanding of vector search and semantic retrieval using FAISS, Pinecone, Weaviate, or Vespa.
- Familiarity with prompt tuning, few-shot learning, zero-shot techniques, and evaluation methodologies.
- Experience with cloud platforms (AWS, GCP, Azure) and MLOps tools for deployment and monitoring.
- Solid grasp of version control (Git), CI/CD pipelines, and containerization (Docker, Kubernetes preferred).