ML / LLM Application Engineer

About the role

TransUnion's AI team seeks an ML / LLM Application Engineer to build production-ready applications powered by large language models (LLMs). You will own end-to-end delivery of LLM-driven features—from design and prototyping to deployment, monitoring, and optimization—focusing on LLM application development, agent orchestration, and system integration rather than deep ML research. This hybrid role is based in Chennai and requires working in-person at an assigned TU office at least two days per week.

What you’ll do

• Design, build, and deploy LLM-powered applications using LangChain, LangGraph, and related tooling.
• Maintain modular, scalable agent workflows for multi-step reasoning, planning, and tool execution.
• Implement and own RAG pipelines, including data ingestion, chunking strategies, embedding workflows, and contextual retrieval.
• Develop effective prompting strategies, tool calling logic, memory handling, and guardrails to ensure reliable system behavior.
• Integrate LLM applications with vector databases to enable semantic search and contextual intelligence.
• Own production considerations such as latency, cost optimization, reliability, and observability.
• Collaborate with product, platform, and data teams to translate business requirements into robust AI solutions.
• Contribute to architectural decisions, engineering best practices, and reusable patterns for LLM-based systems.

What you’ll bring

• 2+ years of experience building production software, with a strong focus on AI-powered or LLM-based applications.
• Strong proficiency in Python for building APIs, backend services, AI workflows, and maintaining ML pipelines and services.
• Solid understanding of machine learning fundamentals, including supervised and unsupervised learning, model evaluation, and feature engineering.
• Hands-on experience with at least one machine learning framework or library (e.g., scikit-learn, PyTorch, TensorFlow).
• Hands-on experience developing LLM applications using frameworks such as LangChain and/or LangGraph.
• Solid understanding of LLM and NLP fundamentals, including embeddings, transformers, vector similarity search, and prompt design.
• Practical experience implementing retrieval-augmented generation (RAG) systems.
• Experience integrating vector databases (e.g., FAISS, Pinecone, Weaviate, or similar).

Nice to have

• Experience working with cloud platforms such as AWS, Azure, or Google Cloud.
• Familiarity with distributed systems, async processing, or microservice architectures.
• Exposure to MLOps practices, including model evaluation, experiment tracking, monitoring, and rollback strategies.
• Experience handling large-scale structured and unstructured data.
• Exposure to multimodal LLM applications (text + image, audio, or video).
• Familiarity with lightweight model adaptation techniques (e.g., inference optimization, prompt tuning, LoRA) is a plus.

Skills