Job Description :

We are building next-generation GenAI capabilities for a fintech platform, and we are looking for a GenAI / Agent Developer to design and ship production-grade LLM applications. You will own the end-to-end lifecycle of Retrieval-Augmented Generation (RAG) pipelines and agentic workflows from prototype to secure, scalable deployment on AWS. This is a hands-on engineering role for someone who has moved GenAI from notebooks into real, reliable products handling sensitive financial data.

Key Responsibilities :

- Design, build, and maintain RAG pipelines including document ingestion, chunking, embedding, retrieval, re-ranking, and grounded generation.

- Design and orchestrate multi-step, multi-tool AI agents that plan, call tools/APIs, and complete complex financial workflows reliably.

- Develop and deploy GenAI solutions on AWS with production standards for scalability, observability, cost control, and security.

- Implement guardrails, evaluation, and monitoring to ensure accuracy, safety, and hallucination control on financial and regulated content.

- Integrate LLM services with internal APIs, vector stores, and knowledge/graph layers.

- Use Claude Code to accelerate development, prototyping, and code review.

- Partner with backend, semantic, and QA engineers to deliver features end-to-end.

- Continuously benchmark models, prompts, and retrieval strategies to improve quality, latency, and cost.

Mandatory Skills :

- Python : strong, production-level proficiency (must-have).

- GenAI / LLM application development : hands-on experience shipping LLM-powered features to production.

- RAG architecture : practical experience designing retrieval pipelines (chunking, embeddings, retrieval, re-ranking, grounding).

- AWS : building and deploying applications on AWS services.

- Claude Code : comfortable using Claude Code (or equivalent AI coding assistants) in daily development.

Nice-to-Have Skills :

- Vector databases (e.g., Pinecone, Weaviate, pgvector, OpenSearch, FAISS).

- Agent orchestration frameworks (LangGraph, CrewAI, or similar).

- Prompt engineering & evaluation tooling (systematic prompt design, offline/online eval, LLM-as-judge, tracing).

- Exposure to LLMOps, model observability, and prompt/version management.

- Experience with regulated or financial data (security, privacy, compliance awareness).

What We Look For :

- A builder who ships : you have taken GenAI features from concept to production, not just demos.

- Strong grasp of LLM failure modes (hallucination, prompt injection, drift) and how to mitigate them.

- Comfort with ambiguity and rapid iteration in a fast-moving product environment.

- Clear communication and collaboration across engineering and product teams.

Fintech Context :

Because this is a fintech platform, awareness of data security, PII handling, and responsible-AI practices for regulated environments is highly valued.