Full Stack AI Engineer - Agent Prototyping & Development

Swap; Greater London; On-site; GBP 65,000 - 85,000; Full time

Job summary

A leading technology firm is seeking a Full Stack AI Engineer to develop and deploy innovative conversational AI agents for e-commerce. This role involves creating responsive web applications, prototype scalable shopping agents, and integrating various APIs for enhanced user experiences. Ideal candidates will have 3+ years in AI/ML engineering and a strong background in full-stack development. The position offers an opportunity to work in a dynamic team focused on cutting-edge technologies.

Qualifications

• 3+ years in AI/ML engineering and conversational AI systems.
• Proficiency with frameworks like PydanticAI and OpenAI Agents SDK.
• Experience in full stack development with Python/FastAPI.

Responsibilities

• Design and prototype scalable shopping agents for e-commerce.
• Build evaluation frameworks for shopping journeys and integrate APIs.
• Implement rapid CI/CD pipelines for seamless deployments.

Skills

AI/ML engineering

Agent Framework Expertise

Full Stack Development

Multi-Modal Development

Cloud-Native Deployment

Tools

PydanticAI

LangGraph

CrewAI

OpenAI Agents SDK

Google Cloud Platform

Job description

Full Stack AI Engineer - Agent Prototyping & Development

London

We are seeking a Full Stack AI Engineer to rapidly architect, prototype, and deploy cutting‑edge conversational AI agent systems for e-commerce. You’ll build scalable proof‑of‑concepts at speed, implement multi‑modal AI pipelines, and develop production‑ready systems with modern web interfaces that quickly adapt to emerging agentic technologies.

What You’ll Do

• E‑Commerce Agent Development: Design and prototype scalable shopping agents using frameworks like PydanticAI, LangGraph, CrewAI, and OpenAI Agents SDK with advanced tool calling for inventory queries, price comparisons, order management, store control, and personalised product recommendations
• Multi‑modal Shopping Experiences: Engineer responsive web applications integrating LLM and VLM models for conversational AI experiences, build intuitive frontend interfaces for product image and description analysis, develop voice‑enabled conversational commerce features, and create seamless user experiences across web and mobile platforms
• E‑Commerce Integration & Evaluation: Build evaluation frameworks using tools like LangSmith, Phoenix, Langfuse, or DeepEval with automated testing pipelines for shopping journeys, rapidly integrate e‑commerce APIs (payment processors, inventory management, CRM systems) with secure frontend implementations whilst maintaining security and PCI compliance
• MCP Integration: Implement Model Context Protocol servers for seamless tool integration and standardised agent communication
• Fast Deployment & DevOps: Implement rapid CI/CD pipelines with containerised deployments, automated testing for e‑commerce workflows, real‑time monitoring dashboards for conversion rates and performance optimisation, and maintain both frontend and backend deployment pipelines

Skills & Qualifications

• 3+ years in AI/ML engineering with rapid prototyping experience in conversational AI systems, agent architectures, or fast‑paced ML deployment environments, ideally with e‑commerce exposure and full stack development experience
• Agent Framework Expertise: Proficiency with PydanticAI, LangGraph, CrewAI, AutoGen, or OpenAI Agents SDK, including agent memory, tool integration for shopping APIs, distributed systems design, and e‑commerce workflow orchestration
• Full Stack Development: Python/FastAPI backend development, modern frontend frameworks (React, Vue, or similar), interactive tools like Streamlit/Gradio or modern AI tools such as Bolt.new & Lovable.dev, database design for customer data and inventory, and responsive web design principles
• Multi‑Modal & Infrastructure: Experience building multi‑modal systems with vision‑language models for product catalogues, speech processing pipelines with web integration, MLOps platforms, rapid model serving, and real‑time inference systems with corresponding user interfaces
• Cloud‑Native Deployment: Ideally Google Cloud Platform expertise including rapid Cloud Run deployments, Vertex AI, Vector Search for product recommendations, Pub/Sub for order processing, monitoring, frontend hosting solutions, and Infrastructure as Code

Performance & Evaluation: Experience with system profiling, latency optimisation for real‑time shopping experiences, LLM evaluation pipelines for shopping conversations, automated testing frameworks for both frontend and backend, continuous validation, and web performance optimisation