Project Description:

Designed and developed production-grade LLM-powered enterprise applications using Python, LangChain, and Retrieval-Augmented Generation (RAG) architectures. Built end-to-end AI pipelines incorporating prompt engineering, query rewriting, intent classification, embeddings, vector databases, retrieval, reranking, and response synthesis to deliver accurate, low-latency responses.

Location: Bangalore or Pune

Responsibilities include:

- Design and implement end-to-end LLM application pipelines using modern orchestration frameworks.
- Own prompt engineering, model quality, RAG accuracy and AI safety across multiple AI applications.
- Build and optimize multi-step LLM workflows including intent classification, query rewriting, retrieval, reranking and response synthesis.
- Develop scalable Retrieval-Augmented Generation (RAG) systems with high retrieval precision and low latency.
- Implement production-grade AI guardrails, jailbreak protection, prompt injection mitigation and hallucination reduction strategies.
- Design automated evaluation pipelines for prompts, retrieval quality and model performance.
- Optimize latency, throughput and cost across complex LLM call chains.
- Integrate NVIDIA AI technologies including NIM, NeMo, NeMo Guardrails and NVIDIA Riva.




- Build streaming AI APIs and conversational experiences supporting multi-turn interactions.
- Collaborate with platform, backend and product engineering teams to deliver enterprise AI solutions.
- Monitor production AI systems, identify quality degradation and continuously improve model performance.

Requirements:

- 4+ years of software engineering experience.
- At least 2 years building production LLM-powered applications.
- Expert-level Python development.
- Strong experience with LLM prompt engineering and prompt optimization.
- Experience designing production RAG architectures.
- Experience with LangChain, LlamaIndex or custom orchestration frameworks.
- Solid understanding of embeddings, vector databases and retrieval optimization.
- Experience implementing AI safety, guardrails and prompt injection protection.
- Experience designing automated AI evaluation pipelines (RAGAS, TruLens or equivalent).
- Experience optimizing latency and quality across multi-stage LLM pipelines.
- Experience developing streaming AI APIs.
- Strong understanding of REST APIs and asynchronous Python development.
- Experience deploying AI applications using NVIDIA NIM and NeMo technologies.
- Excellent debugging, analytical and communication skills.
- Strong English communication skills.

📌 6 AI Application Engineer (India)
🏢 Customertimes
📍 India