About the job
FlatGigs is collaborating with a prominent banking client engaged in a transformative Trade initiative powered by Generative AI. We seek innovative engineers who can architect and implement enterprise-grade Agentic AI systems that seamlessly integrate large language models (LLMs) with secure, structured private data environments.
This is a production-focused engineering role within regulated enterprise systems, not a research position.
Immediate joiners are highly preferred.
Role Overview
The AI Engineer will be tasked with designing, building, and deploying sophisticated multi-agent AI systems within enterprise trade platforms.
The ideal candidate will possess robust full-stack engineering skills alongside extensive expertise in:
- Agentic AI architectures
- LLM orchestration
- Secure enterprise integrations
- Prompt engineering
- Vector search systems
Key Responsibilities
Agentic AI & LLM Architecture
- Design and implement comprehensive multi-agent workflows
- Integrate LLMs with structured private enterprise data
- Develop modular AI agents capable of reasoning and tool orchestration
- Ensure compliance with explainability and traceability requirements in regulated environments
Prompt Engineering & Reliability
- Create and optimize advanced prompt frameworks
- Establish guardrails and fallback mechanisms for enterprise stability
- Develop evaluation frameworks to assess hallucination and output consistency
Full Stack Engineering (Node.js / React)
- Build scalable backend services using Node.js (Express / NestJS)
- Develop secure frontend applications with React and TypeScript
- Create RESTful APIs to expose AI workflows
- Embed AI modules into enterprise trade systems
Secure Enterprise Integration
- Integrate Generative AI services using secure API frameworks
- Implement role-based access control
- Ensure compliance with sensitive data handling standards
AI Tooling & Frameworks
- Utilize LangChain / LangGraph for multi-agent orchestration
- Implement memory layers and workflow graphs
- Work with vector databases (Redis preferred)
- Design and optimize RAG (Retrieval-Augmented Generation) pipelines

