AI Agent Developer

Vallum Associates; City Of London; On-site; GBP 120,000 - 180,000; Full time

Job summary

Vallum Associates in London is hiring an AI Agent Developer to own the technical delivery of an enterprise AI agent platform that enables task automation for Finance Operation users. You will combine architectural design with hands-on full-stack coding to build secure, scalable services and experiences for developers and business users.

You will integrate LLMs, AI agents and knowledge augmentation, from requirements through coding, testing and orchestration, applying modern DevSecOps practices

Qualifications

• 8+ years in backend/platform engineering or enterprise delivery.
• Experience delivering AI solutions into production in a commercial environment over the past three years.
• Strong system thinking with a focus on resilience, scalability, high availability, maintainability and performance.

Responsibilities

• Lead architecture and implementation of an AI agent platform (hands-on).
• Translate model capabilities into reusable components and SDKs.
• Embed AI capabilities into business workflows and establish quality/measurement practices.
• Deliver production-grade cloud-native services with CI/CD, DevSecOps and observability.

Skills

Python

React

System design

Architecture judgement

Performance

Security best practices

Tools

LangChain

LangGraph

Claude Code

Codex

Cursor

Copilot

Kubernetes

CI/CD tooling

Google Cloud

Job description

Role Summary

We’re hiring an AI Agent Developer to own the technical delivery of an enterprise AI agent / AI-native engineering platform that supports task automation for our Finance Operation business users. You’ll combine architectural design with hands‑on, full‑stack engineering, writing production code to build secure, scalable services and user experiences that developers and business users will adopt.

You’ll integrate large language models (LLMs), AI agents and knowledge augmentation into the solution design, from requirements and design through coding, testing and environment orchestration. You’ll apply modern full‑stack engineering practices: microservices, micro frontends, continuous integration and continuous delivery (CI/CD), integrated development, security and operations (DevSecOps), observability and cloud‑native delivery.

Key Responsibilities

1) AI Agent Platform Architecture & Delivery (hands‑on)

• Lead architecture and implementation of an AI agent platform, including task orchestration, tool invocation, context/session management, memory, and multi‑agent collaboration.
• Translate emerging model capabilities into reusable platform components that teams can adopt safely and quickly, including software development kits (SDKs), services, templates and guardrails that constrain AI behaviour.
• Drive real implementation in complex engineering scenarios, not just prototypes—optimised for reliability, latency, and cost.

2) AI‑native Engineering Workflow Enablement

• Embed AI capabilities into Business workflows, covering:
  • AI‑assistant agent
  • AI evaluation
  • Natural Language to SQL generation
• Establish quality and measurement practices for AI features (accuracy, coverage, drift, prompt/tool regression, cost).

3) Knowledge Augmentation (RAG / GraphRAG / Knowledge Graph)

• Design and implement a knowledge layer that grounds AI responses in enterprise information and improves their correctness, covering:
  • RAG, GraphRAG, and knowledge graph design/integration
  • indexing strategies, retrieval evaluation, and relevance tuning
  • governance for enterprise knowledge usage (permissions, lineage, auditability)

4) Engineering Delivery Excellence (Full Stack + Microservices + DevOps)

• Stay hands‑on delivering production‑grade full stack solutions (Python backend microservices/ React frontend UI), applying secure engineering practices and strong system design.
• Build and operate cloud‑native services with CI/CD, DevSecOps, test automation, environment management, and robust observability (metrics/logs/traces), driving high availability and operational readiness.
• Collaborate across Product/Engineering/QA/Operations to standardise, productionise and continuously improve platform capabilities; mentor engineers and uplift delivery maturity using practical engineering metrics (e.g., DORA).

Required Experience

8+ years in backend/platform engineering, architecture, or similar enterprise‑scale delivery roles.

Experience working in a commercial environment over the past three years, delivering AI solutions into production.

System thinking: Strong ability to design and evolve complex systems with a focus on resilience, scalability, high availability, maintainability, and performance.

Engineering strength:

• Proficiency in Python and React, supported by strong coding fundamentals, architectural judgement and hands‑on experience delivering complex systems; Java proficiency may also be required.
• Strong hands‑on skills in Microservice development, Kubernetes, CI/CD and quality automation

AI/Agent Architecture Expertise

• Strong understanding of LLM/agent architecture patterns:

• task decomposition, planning, tool use/function calling
• context management and memory design
• execution loops, failure handling, and safety guardrails

• Hands‑on experience with LangChain and/or LangGraph (or equivalent orchestration frameworks).
• Familiarity with AI developer tools such as Claude Code, Codex, Cursor, Copilot.

Full Engineering on Cloud

• Micro service design and development for resilient and reliable system with excellent skillset on database, message broker and CI/CD toolkits
• Cloud-native engineering experience (e.g., Google Cloud) and platform mindset.
• Certifications are a plus (e.g., GCP, architecture/programming certs).
• Knowledge/certification in Kubernetes, and ecosystem tools such as Message Broker e.g. GCP pub/sub, Spring Gateway (or equivalents) is a plus.
• Experience contributing to engineering communities is a plus.

What Success Looks Like

A secure, scalable AI agent platform running reliably in production with clear adoption across teams.

Measurable engineering impact (delivery speed, quality, stability) backed by strong telemetry and feedback loops.

Well-governed knowledge augmentation (RAG/GraphRAG/knowledge graphs) that improves accuracy and trust.

Efficient, end‑to‑end delivery of production‑ready solutions through hands‑on full‑stack engineering.

A team culture that ships quickly, learns continuously, and operates with blameless accountability