AI & Data Engineer

Remote (UK-based) | Full-Time

Reactive Markets is the 2026 OTC Trading Platform of the Year (Risk.net). Our network handles over $50 billion in daily trading volumes across FX, Equities, and Cryptocurrency, connecting 40+ of the world's leading liquidity providers.

We are investing heavily in AI and data as core strategic priorities for 2026 and beyond. Our execution analytics differentiate us with clients, and our internal AI initiative — Project Iris — is already changing how we investigate production issues, support clients, and run our platform.

This is a platform engineering role. As in every engineering role here, the product end to end comes first: how the platform behaves, how it fails, who depends on it. What this seat owns is the data backbone — the data plane nearly everything else is built on, from the calculations the business bills and advises on to the analytics clients see and the AI tooling layered above. Everything here is about data; AI sits on top of it.

It is the next phase, not a rewrite. Our Python estate grew up alongside the analytics it supports, and this role gives it real shape and structure — a shared foundation holding our authoritative business calculations, markouts and spreads through to fees and instrument mappings, each implemented and tested once and then called from notebooks, scheduled jobs, APIs and AI tooling alike — before building it out substantially as that backbone grows. Python is the primary language, with cross-training into Go; first-language Python is what we would most like to find.

Your closest customer is our Liquidity Management desk, whose execution analytics differentiate us with clients: you will engineer the analysis they have come to depend on, while leaving their exploratory work fast and unconstrained. Your closest engineering partners are our ClickHouse engineers, with whom you will build robust, scalable data processing pipelines over the plant they own. You do not need to have built LLM applications before.

• The data backbone — the next generation of the shared foundation for our business rules and calculations, made for the people who depend on it: versioned, tested, safe to upgrade

• Research to production — engineering the analysis the business depends on into tested, monitored pipelines, with the analyst still free to explore

• Data processing pipelines — built with our ClickHouse engineers, robust and scalable over a plant capturing billions of rows a day, with the reconciliation that proves a figure is right

• Analytics and monitoring — real-time anomaly detection and platform-health monitoring (our "Radar" initiative), and performant queries over very large datasets

• AI tooling, agents and MCP servers — investigation tooling that turns hours of manual trawling into guided workflows, and guardrailed access that lets agents act on real systems safely

Essential:

• End-to-end systems thinking — you reason about the whole system, not just your part of it: how it behaves under load, how it fails, what depends on it, and who is affected when it is wrong

• Strong Python — production Python that other people depended on, with real views on packaging, dependency management, testing and typing. First-language Python is what we would most like to find

• The ability to work with a business function — evidence you have embedded with domain experts who are not engineers, learnt their world, and earned the right to change how things work

• SQL — comfortable writing analytical queries against large datasets (ClickHouse, PostgreSQL or similar)

• Git and Linux fundamentals

Important:

• Library and API design — versioning, backwards compatibility, and designing from the caller's side

• Production discipline — tests, monitoring, reproducibility, and fixing causes rather than symptoms

• Docker, Kubernetes and AWS — your work runs in production infrastructure

Welcome, and we will help you grow it:

• Go — cross-training is expected; prior experience is a bonus, not a requirement

• LLM APIs and tooling — Claude, OpenAI or similar; prompt engineering, tool use, MCP

• TypeScript / JavaScript — useful for Slack integrations and web interfaces

• Financial services experience — a plus but not essential. Domain knowledge can be learned; engineering instinct cannot

How you work:

• You ship. You iterate quickly, get things in front of users, and improve based on feedback

• You respect other people's expertise. The analyst who wrote the notebook knows something you do not. Your job is to make their work durable, not to correct their taste

• You own it. When you build something, you take responsibility for its behaviour in production — you monitor it, you fix it, you make it better

• You are honest when a number is wrong. You tell the people who acted on it before you quietly correct it

• You write things down and communicate. We are distributed across time zones, and documentation is how we scale knowledge — every well-written page also makes our AI tooling better

• You embrace AI as a tool. You use it to amplify your own work, and hold generated code to the same standard as anything else

• You own the backbone — the data plane you shape carries the analytics clients see and the figures the business runs on, and everything built after it inherits your decisions

• A rare vantage point — engineering alongside a trading desk. You will learn how the market actually works, from the people who trade it

• Ownership — small teams, clear accountability, direct impact on the platform and our people

• An excellent team — engineers who care about craft, collaborate openly, and hold each other to high standards

• Competitive package — up to 30 days leave + bank holidays, private health insurance, pension, life insurance, dental/optical cashback, cycle to work, socials and offsites

• Remote-friendly, at a sustainable pace — work from home with flexibility, and we don't burn people out. Balance is how we stay sharp

• Growth — Python, data, AI and platform engineering in one role, with the scope growing as the company scales

We believe in transparency, honest feedback, and writing things down. We celebrate delivery, not activity. We frame AI as empowering people — removing toil, amplifying capability, enabling higher-value work.

We keep our process focused and respectful of your time. Our process follows three stages.

Stage 1: Initial Conversation with Talent Acquisition
Format: Video call, ~30-45 minutes

Stage 2: Hiring Manager Conversation with the relevant team lead or hiring manager
Format: Video call, ~60 minutes

Stage 3: Technical Interview with two members of the relevant engineering team
Format: Video call, interactive session. ~60 minutes

What to expect in each stage will be explained by Talent Acquisition if you're invited to interview. Throughout the process, we aim to keep momentum with no unnecessary delays between stages, and feedback typically comes within a few business days.

*Please note that depending on availability, Stage 2 and 3 may swap around.

Please apply directly via this job posting — all applications, including those submitted via LinkedIn, are managed through our applicant tracking system, so you'll always land in the same place regardless of where you found this role.

Right to Work: Candidates must have the existing right to work in the UK. We are not currently able to offer visa sponsorship for this role.

Equal Opportunities: Reactive Markets is an equal opportunities employer. We welcome applications from all qualified candidates regardless of age, disability, gender reassignment, marriage or civil partnership, pregnancy or maternity, race, religion or belief, sex, or sexual orientation.

Reasonable Adjustments: If you need any adjustments or support during the application or interview process, please let us know — we're happy to accommodate.

Data Protection: By applying, you consent to Reactive Markets processing your personal data for recruitment purposes, in line with our privacy policy [link to be added].

#Hiring #SoftwareEngineering #AI #DataEngineering #FinTech #Python #FullStack #RemoteWork