GenAI Engineer

Experience: 3+ years of overall experience

Primary Skills (Must-Have): Python, RAG, SQL (MySQL/PostgreSQL/MSSQL/RDS/Aurora), LLM (GPT/Claude/Gemini/Llama), Agentic AI (LangGraph/LangChain/MCP), Vector DB (FAISS/Weaviate/Pinecone/Chroma/pgvector), AWS (CodePipeline, S3, EC2, ECS, Lambda, SageMaker)
Secondary Skills (Good-to-Have): Integrations, Snowflake

Location: Bengaluru (Hybrid)
Shift (IST Hours): General Shift

About the Role

We are looking for a GenAI Engineer to build enterprise grade chatbots and intelligent assistants powered by the latest Large Language Models. You will work on real finance data, where a confident but wrong answer is not acceptable, so accuracy, grounding and explainability sit at the heart of everything this team builds. This is a hands on engineering role. You will take GenAI ideas beyond the prototype stage and turn them into reliable, production ready systems that run at scale on AWS, working closely with the Product, Data and Platform teams.

What You Will Do

- Build intelligent chatbots: Design and develop LLM-powered assistants using OpenAI, Anthropic and other leading models, combining Retrieval Augmented Generation (RAG), tool calling and agent frameworks such as LangChain and LangGraph.
- Design agentic workflows: Create Agent-to-Agent (A2A) architectures where specialised agents hand off tasks to each other, and use MCP to connect agents to tools, APIs and databases for multistep reasoning and autonomous workflows.
- Engineer retrieval pipelines: Build the chunking, embedding and retrieval layer on vector databases, using hybrid search (semantic plus keyword) and cross encoder reranking so the model always sees the most relevant context.
- Reduce hallucinations: Apply grounding, reranking,



source citations and confidence scoring so that every answer can be traced back to its source, and low confidence answers are flagged instead of guessed.
- Work with finance and enterprise data: Handle structured and unstructured financial datasets with a strong focus on accuracy, data governance and secure access.
- Evaluate and improve quality: Set up offline evaluation on test datasets before release and online evaluation on live traffic after release, along with prompt and version management.
- Deploy and monitor at scale: Ship AI services on AWS using cloud native and MLOps practices, build CI/CD pipelines for AI and inference services, and monitor performance, cost and reliability in production.
- Collaborate across teams: Partner with Product, Data and Platform teams to understand business needs and deliver AI solutions that people trust and actually use.

Must Have Skills

- Python: 3+ years of hands on development experience
- SQL: MySQL/PostgreSQL/MSSQL/RDS/Aurora
- LLMs: Working with GPT/Claude/Gemini/Llama in real applications
- RAG: Proven experience designing and shipping retrieval-based solutions
- Vector Databases: Experience with FAISS/Weaviate/Pinecone/Chroma/pgvector
- Agentic AI: Hands-on experience with LangGraph/LangChain/MCP
- AWS: Experience with CodePipeline, S3, EC2, ECS, Lambda and SageMaker

Positive to Have

- Experience with Snowflake and DBT for data pipelines
- Experience integrating AI systems with enterprise applications and APIs
- Exposure to LLM evaluation and observability tools such as LangSmith, MLflow or Weights & Biases
- Prior work on AI solutions in the finance or banking domain

Why You Should Apply

- Build GenAI systems that go live and are used at enterprise scale, not just demos.
- Work on the current generation of the agentic AI stack, including LangGraph, MCP and A2A.
- Solve meaningful problems in finance, where accuracy and trust really matter.

📌 GenAI Engineer (Bengaluru)
🏢 Carnera Technologies
📍 Bengaluru