Applied AI Engineer (up to £180k)

Dex; Greater London; On-site; GBP 120,000 - 170,000; Full time

Job summary

Dex is seeking a backend engineer to design production-grade, autonomous infrastructure powering an acquisition engine. You will craft memory, tool-calling, orchestration, and evaluation layers enabling AI agents to handle live workflows across landlords, tenants, and service providers.

You will ship robust systems beyond simple API wrappers, with a focus on reliability, observability, and scalable architecture that supports rapid onboarding of acquired businesses.

Qualifications

• Track record of shipping LLM-powered or agentic systems into production environments.
• Strong backend engineering foundation in TypeScript/Node.js, or deep Python experience.
• Demonstrated architectural judgement on where to deploy probabilistic LLM agents and deterministic logic.

Responsibilities

• Architect the core platform for multi-agent orchestration, context management, tool integration, and state persistence.
• Build automated evaluation harnesses and observability tooling to benchmark agent reliability.
• Standardise operational processes into modular, reusable agentic components for quick onboarding.

Skills

Backend engineering

TypeScript/Node.js

Python

LLM-powered systems

Architectural judgement

AI coding agents

Job description

This role is with one of Dex's trusted partner companies. We work closely with their teams to truly understand their culture, goals, and what they're looking for, so we can match you with the right opportunity and give you context about the role before you commit to a process.

Dex is an AI recruiter agent that helps you run your job search. Tell Dex your stack, seniority, and what you want to build. We will manage your applications and surface other opportunities that are a fit.

The role

Most property tech startups fail because selling SaaS into fragmented, legacy agencies is a slow grind. This company takes the opposite approach: they acquire independent operators outright and rebuild the entire operational layer on their own proprietary software. Backed by top-tier global VCs and growing rapidly through acquisitions, they now manage thousands of properties and turn manual legacy operations into automated, reliable software systems at scale.

You will build the core agentic infrastructure that powers this acquisition engine. That means designing the memory, tool-calling, orchestration, and evaluation layers that allow AI agents to handle live workflows across landlords, tenants, and service providers. This is not an exploratory prompt-engineering role, and it is not building a bolted-on conversational chatbot. You are engineering robust, production-grade autonomous infrastructure where hallucinations cause real-world operational failure, turning custom agency processes into reusable platforms that launch in days.

The work

• Architect the core platform for multi-agent orchestration, context management, tool integration, and state persistence across operational workflows.
• Build automated evaluation harnesses and observability tooling to benchmark agent reliability and catch regressions before systems go live.
• Standardise custom operational processes into modular, reusable agentic components so newly acquired businesses can be onboarded in days.
• Define the architectural boundary between autonomous agent execution, deterministic backend code, and human intervention safeguards.
• Leverage agentic software engineering practices and coding agents as a core part of your daily development cadence.

What You Bring

• Track record of shipping LLM-powered or agentic systems into real production environments, well beyond basic API wrappers, proof-of-concepts, or hackathon demos.
• Strong backend engineering foundation in TypeScript/Node.js, or deep Python experience with the ability to ship in a TypeScript-first codebase.
• Demonstrated architectural judgement on where to deploy probabilistic LLM agents, where to enforce deterministic logic, and where human-in-the-loop review is mandatory.
• Active, daily use of AI coding agents to decompose tasks, write code, and accelerate your own software engineering workflow.

Why apply through Dex

Most applied AI postings are either trivial API wrapper jobs or disconnected R&D experiments with no path to production. Dex cuts through the noise, matching you directly with high-conviction teams building real agentic infrastructure behind closed doors. You get fully briefed on the company, architecture, and team before you speak with them, skipping the cold-apply queue entirely.

As part of the recruitment process at Dex, we process your personal data in accordance with our Privacy Notice for Job Applicants. This notice explains how and why your data is collected and used, and how you can contact us if you have any concerns.