MLOps Platform Developer / Full-Stack AI Engineer

Job Title: MLOps Platform Developer<br />Location: London<br />Salary: Depending on experience<br />Job Type: Full time, Permanent<br />This is a rare opportunity to be first in the door to continue the development of our engineering platform.<br />You will own a production system end to end - the web application, the database estate, the data pipelines, the deployments, the integrations and the LLM serving stack.<br />What's involved?

• Genuine full-stack breadth in one codebase: a React/TypeScript application, a Postgres estate with 100-million-row telemetry tables, serverless functions, MQTT ingestion from building management systems.
• LLM model training in the specialist fields our clients value most. It could be a housing estates network or a shopping centre, each LLM is tailored to their exact use case.
• Neural LLM-assisted engineering at the sharp end: coding agents are a first-class part of how the platform is built, and you will get very good at directing them, verifying them and knowing where they lie.
• Operational technology: BACnet, Trend, JACE and Tridium controls, data points all working together to optimise and commission real engineering systems.
• The energy transition, concretely: heat networks are central to UK decarbonisation and growing fast; the domain knowledge you build here is scarce and compounding.
• Small-company reality: direct access to clients, sites, commercial decisions and the CEO, we are a bureaucracy free organisation with a. clear delivery and results focus.<br />What you will do:
• Operate the LLM estate: run LoRA fine-tuning cycles and evaluation gates on our GPU hardware, manage vLLM serving (including multi-adapter deployments) alongside production, promote or roll back model versions on the gate results, and keep the serving watchdogs healthy. The CEO retains model direction; this role owns the running of the pipelines.
• Develop, deploy and maintain the web application (React, TypeScript, Vite) and its offline-first engineer mobile surface, with serverless functions on Vercel and Database including the email-ingestion and messaging pipelines.
• Own the data layer: Postgres on Database - schema design, row-level security for tenant isolation, SQL functions, materialized views, scheduled jobs, query performance at telemetry scale.
• Run the telemetry pipeline: MQTT ingestion from site BMS equipment, collector services, data-quality rollups, retention, and the monitoring that guards them.
• Test and verify to the house standard: every change exercised end-to-end against production, the automated invariants suite kept green and extended with each feature.<br />About you:<br />Qualifications and experience:<br />Essential:
• Degree in computer science, software engineering or a related discipline, or demonstrably equivalent professional experience.
• ML operations capability: comfortable operating model-serving infrastructure and running established training and evaluation pipelines on Linux/GPU hardware - process management, resource monitoring, versioned promote/rollback. Training pipelines exist and are documented; you must be able to run, monitor and recover them, not design them from scratch.
• Five or more years building and running production web applications full-stack, with sole or lead ownership of at least one system in live use.
• Strong TypeScript and React; strong SQL and relational schema design on PostgreSQL, including performance work on large tables.
• Serverless or cloud deployment experience (Vercel, AWS Lambda, Cloudflare Workers or similar) and comfort operating what you ship: logging, monitoring, incident response.
• API integration experience against third-party systems (REST, OAuth2/JWT, webhooks), including at least one messy real-world system of record.
• Python for scripting, data processing or pipeline work.<br />Preferable:
• Deeper LLM experience: fine-tuning methodology, corpus design, evaluation-harness construction, vLLM internals, or heavy daily use of AI coding agents.
• ERP, FSM or works-management domain experience - job lifecycles, scheduling, SLAs, parts, timesheets, invoicing - or heat-network / building-services domain experience (BMS, metering, HIUs, energy-centre plant).
• IoT or telemetry experience: MQTT, time-series data, meter or sensor fleets; BACnet or Trend/Tridium controls exposure a bonus.
• Supabase specifically (RLS policies, edge functions, realtime), and Deno.
• Experience in a small company or founder-adjacent role where you owned outcomes rather than tickets.<br />Notice to Agency and Search Firm Representatives: We are not accepting unsolicited CVs from agencies and/or search firms for this job posting. CVs submitted to any employee by a third-party agency and/or search firm without a valid written &amp; signed search agreement, will become our sole property. No fee will be paid if a candidate is hired for this position as a result of an unsolicited agency or search firm referral. Thank you.<br />Only applicants who are legally entitled to work in the UK and are currently resident in the UK are invited to apply.<br />Please click on the APPLY button to send your CV for this role.<br />Candidates with the experience and relevant job titles of: Lead Full-Stack Engineer, AI Application Engineer, Applied AI Engineer, LLM Ops Specialist, AI Platform Engineer, MLOPS Engineer, Founding Engineer, Full-Stack AI Developer, AI Software Developer may also be considered for this role