Forward Deployed Engineer

This role focuses on designing, building, and deploying production-grade Agentic AI and multi-agent systems within enterprise environments. You will work across AI engineering, backend development, cloud deployment, and production delivery.

Design and build multi-agent AI systems using LangGraph, LangChain, CrewAI, or similar orchestration frameworks.

Develop end-to-end RAG pipelines, covering chunking strategies, embeddings, retrieval optimisation, reranking, and hallucination evaluation.

Build prompt frameworks, agent role instructions, escalation flows, and guardrail mechanisms for enterprise AI applications.

Implement AI security and guardrails including PII redaction, input sanitisation, output validation, adversarial testing, and data protection controls.

Develop production-grade backend services and APIs using Python and FastAPI, including streaming APIs, authentication layers, and enterprise integrations.

Deploy AI workloads using AWS services such as Bedrock, SageMaker, ECS, EKS, API Gateway, Cognito, CloudWatch, IAM, and Secrets Manager.

Build CI/CD pipelines, automated testing workflows, infrastructure automation, observability tooling, and production monitoring frameworks.

Define testing and evaluation strategies across unit testing, integration testing, retrieval testing, adversarial testing, and AI evaluation.

Contribute to architecture reviews, low-level designs, technical documentation, and reusable engineering assets.

Lead technical workstreams and provide technical guidance to junior engineers where required.

Experience in software engineering, AI engineering, cloud engineering, or backend platform development.

Strong Python engineering experience with a background in building production-grade backend systems.

Hands-on experience building AI applications using LLM APIs, RAG systems, orchestration frameworks, or multi-agent workflows.

Experience with LangGraph, LangChain, CrewAI, or similar Agentic AI frameworks.

Strong understanding of retrieval systems, embeddings, vector databases, reranking, chunking strategies, and AI evaluation.

Experience deploying applications on AWS, ideally using services such as Bedrock, SageMaker, ECS/EKS, API Gateway, and related services.

Understanding of enterprise AI risks including hallucinations, prompt injection, data leakage, and AI guardrails.

Experience working directly with delivery teams, architecture teams, or enterprise stakeholders would be advantageous.