Remote Senior Applied Machine Learning Engineer - Catalogue Intelligence in Northampton

Who are OnBuy? OnBuy are an online marketplace who are on a mission of being the best choice for every customer, everywhere. We have recently been named one of the UK's fastest-growing tech companies in the Sunday Times 100 Tech list. All achievements we are very proud of, but we don't let that go to our head. We are all laser focused on our mission and understand the huge joint effort ahead of us needed to succeed.

Working at OnBuy: We are a team of driven and motivated people who thrive when working at pace. To succeed at OnBuy you need to take charge and fully own your responsibilities, rolling your sleeves up when needed to 'get it done'. Working at OnBuy you are surrounded by so much opportunity, but you must possess the ability to stay focused and prioritise ruthlessly. Most importantly, you will thrive in an ever-changing environment as we are constantly evolving. At OnBuy, you're not just a number or another cog in a machine. We are creating something really special, and you have the opportunity to affect meaningful change and have your voice heard. We are a close team, who have the opportunity to learn and grow as OnBuy evolves.

About the Role: We’re building a more intelligent, scalable product catalogue across multiple markets. Core capabilities like auto-categorisation and brand detection already exist, but they are not yet connected into a system that consistently drives quality, discovery, and growth. This role owns that system. The Senior Applied ML Engineer - Catalogue Intelligence is responsible for building the machine learning systems that power OnBuy’s catalogue decisioning engine. Working in partnership with the Head of Seller Solutions, who defines catalogue rules and commercial logic, you will design and deploy production-grade systems that automatically improve:

• Product categorisation
• Product data quality and completeness
• Pricing competitiveness insights
• Catalogue coverage and selection
• Product discoverability

This is a hands-on, production-focused role where outputs directly modify the live catalogue and materially impact GMV, conversion performance, search, and discovery.

Core mission: Turn catalogue rules and commercial logic into automated, data-driven systems that continuously improve discovery, data quality, pricing competitiveness, and revenue outcomes.

What you’ll be responsible for: You’ll take ownership of how product data is structured, validated, and used across the platform. This includes:

• Improving how we classify and understand products at scale
• Raising the overall quality of catalogue data and defining what “good” looks like
• Ensuring product data supports effective search, filtering, and discovery
• Identifying gaps in our catalogue and surfacing opportunities for growth
• Improving how our catalogue performs across external channels

You’ll build and evolve the systems and decision logic that enable this, and iterate based on real-world performance and data.

You’ll work across:

• Structured data (catalogue attributes, GTINs, taxonomy)
• Unstructured data (text and images)
• Behavioural data (search, clicks, conversions)

How you’ll work: You’ll build directly using SQL and Python on top of:

• BigQuery
• Airbyte
• Google Datastream

You’ll be working across data pipelines, information extraction, and production ML systems, combining rules, heuristics, and ML/LLMs where appropriate. The focus is on shipping practical systems quickly, validating them with real data, and improving them over time. You’ll work closely with engineering, product, and analytics, but you’ll be expected to own and deliver the core logic yourself. This role is not focused on research or offline modelling. You’ll be expected to build systems that operate in production and directly influence how products appear and perform on the platform.

Experience & Skills: Experience building and shipping production data or ML systems with measurable business impact. Strong Python and SQL skills, with the ability to work across data pipelines end-to-end. You should be comfortable applying modern approaches such as LLMs, multimodal models, and information extraction techniques, and taking them from experimentation into production with proper evaluation, monitoring, and cost control. Experience working with messy, unstructured or semi-structured data (e.g. text, images, product data). Ability to design systems that make decisions, not just predictions. Strong judgement in balancing accuracy, risk, and business impact. Experience with ecommerce or marketplace catalogues is a plus, but not required.

How you’ll operate: Take ownership of problems end-to-end, from idea through to production impact. Build systems that are scalable, testable, observable, and auditable. Design automation with confidence thresholds, monitoring, and feedback loops in mind. Maintain clear documentation, evaluation frameworks, and versioning for models and logic. Work pragmatically, favouring simple solutions that deliver impact quickly. Communicate trade-offs clearly to technical and non-technical stakeholders. You should be comfortable working in environments where data is incomplete, inconsistent, and constantly evolving. You’ll be working across cloud-based data systems (GCP), building and deploying data pipelines and production ML workflows. Experience with orchestration tools (e.g. Airflow or similar), CI/CD, and model deployment practices is beneficial.

The salary on offer for this role is £65000- £75000 depending on experience. We also offer the following benefits: Company Equity- In return for helping us to grow, we’ll offer you company equity, meaning you own a piece of this business we are all working so hard to build. 25 days annual leave +