We are looking for a Junior to Mid-level Analytics Engineer who wants to grow into someone who can do three things equally well: build and maintain the data models that reporting sits on, work with people to understand what they actually need, and use AI-assisted tooling as a normal part of how the work gets done. You don't need to arrive doing all three confidently - you need to be strong in analysis and reporting today, and genuinely keen to develop into the modelling and engineering side with support from the team.
You will be part of the bridge between our small data team and the rest of the organisation. That means turning agreed requirements into reliable, well-modelled data and the reports that sit on top of it, working with our Impact & Insights colleagues who own metric definitions and outcome frameworks rather than duplicating that work.
The majority of the role sits on the platform side - building and maintaining the Bronze/Silver/Gold model alongside the Data Engineer, restructuring the data lake, and documenting what exists. Alongside that, you will support requirements and reporting work with business teams, so that what gets built reflects what people actually need. You will contribute to the team's data models, documentation and governance as we consolidate two organisations onto a shared Microsoft Fabric platform - and you will do it using the AI-assisted tools and workflows the team has adopted as standard, under a peer-review safety gate, which is also where a lot of your own learning will happen.
You will report to the BI Lead and work closely with our Data Engineer, who you'll pair with regularly as you build up the modelling and engineering side of the role. The team is small and deliberately focused - we aim to do fewer things better, rather than spread thin across every request.
Duties and Responsibilities
Data modelling and engineering (primary focus)
- Start contributing to work in the Fabric Lakehouse alongside the Data Engineer - supporting Bronze/Silver/Gold transformations before progressing to building and refactoring them independently.
- Write documented, testable logic for the pieces you own, growing into more complex transformation work as your confidence builds.
- Learn dimensional modelling in practice (fact/dimension design, star schemas) through paired work with the Data Engineer, rather than being expected to make these calls solo from day one.
- Help identify manual Excel-based consolidation processes across departments as candidates for automation, working with the Data Engineer to replace them.
Requirements and business partnering (supporting the wider process)
- Support requirements-gathering sessions with teams across Felix - including finance, fundraising operations, operations and distribution, food sourcing and digital marketing - working within the agreed triage process and alongside the Impact & Insights team rather than in parallel with it.
- Help translate agreed business requirements into clear, achievable technical briefs, learning when to flag constructively where requests outpace available data or team capacity.
- Maintain visibility of what is being built, for whom, and why - supporting the team as a technical point of contact for reporting and data requests routed through the agreed triage process.
- Help manage expectations around timelines, data limitations and priorities, communicating changes clearly and proactively.
Reporting and dashboards
- Design, build and maintain Power BI dashboards and reports that are accurate, well-documented and genuinely adopted by the teams they serve.
- Ensure reports are built on solid, governed data foundations rather than ad-hoc workarounds.
- Implement agreed KPI definitions consistently across models and reports, and keep the shared data glossary up to date as definitions are signed off.
AI-augmented delivery
- Use AI-assisted tooling as a normal part of the workflow from day one - Copilot-style code and DAX assistance, LLM-assisted documentation, and MCP-based agentic tools - with the team showing you how, in line with the permissive-but-reviewed approach.
- Develop judgement about when not to trust generated output: you'll be taught how to verify a model, a measure or a piece of documentation an AI tool produced, and everything goes through peer review before it reaches production, stakeholders or decisions - this is part of how you'll learn, not just a safety check.
- Contribute observations to the team's shared practice around AI-assisted work as you build experience with it.
- Work AI tools iteratively rather than one-shot - refine a brief, check the output against the source data, loop back and improve it, rather than taking the first draft a model gives you and moving on. This applies whether you're writing DAX, drafting documentation or preparing a stakeholder brief.
Data quality and governance
- Flag data quality issues as you encounter them and trace them to their source, working with the Data Engineer where needed.
- Document your models, reports, calculations and assumptions clearly so that others can understand, challenge and build on your work.
- Support consistent naming conventions, metadata practices and ownership models as part of the team's broader governance effort.
Continuous improvement
- Participate in regular team reviews of the Fabric environment, identifying opportunities to consolidate models, reduce compute costs and improve performance.
- Stay curious about what teams are doing with data and proactively surface opportunities where better modelling or analysis could support the charity's mission.
- Keep pace with new AI tools and model capabilities relevant to our stack, try them out, and bring back to the team what's actually worth adopting.
This section must be kept and filled in and do not alter title.
Guidance: Explain what you are looking for in your ideal candidate
This is your opportunity to set out a checklist of the skills, both technical and interpersonal, and the experience you expect applicants to possess.
Personality traits and soft, interpersonal skills that would help them succeed
It should be clear enough that anyone reading the job specification who is considering applying can easily work out whether or not they are suitably experienced or the right type of person to carry out the job.
Essential Criteria
- SQL - the core day-to-day skill for this role. Comfortable writing and reading queries across relational databases, including joins and aggregation, and ready to build up to more complex transformation work.
- Dimensional modelling - you understand fact and dimension design, star schemas, keys and granularity, and what goes wrong downstream when they are off. We do not expect years of it, but you should be able to reason about a model rather than only consume one.
- Python - working exposure to data manipulation (pandas, PySpark or similar). You will use it here from early on, so some practical experience is needed even if it is not your strongest language.
- Power BI - able to build reports and dashboards from existing data models, and to interrogate why a visual looks wrong and trace it back to the model rather than patching the visual.
- DAX - comfortable writing straightforward measures and adapting existing ones; ready to build up to more complex time intelligence and row-context work with support.
- Clear communication with non-technical colleagues - you can explain a data limitation, a definition or a delay in plain language, and understand a requirement well enough to build the right thing.
- Attention to detail - you notice when numbers don't add up and you follow the thread until you understand why.
- High agency - when a tool, a dataset or an approach doesn't give you what you need first time, you try it a different way rather than shelving it or waiting to be told what to do next.
- AI fluency and judgement - you use Copilot-style, LLM and agentic (MCP-based) tools as a normal part of the workflow, and you work them properly: refining a prompt, checking the output, and going again rather than accepting the first answer. You have the humility to have that output peer reviewed.
- Collaborative working style - this team operates transparently and expects you to flag blockers early, share work in progress, and ask questions without hesitation.
- Appetite to grow into engineering - the single most important thing we are looking for. The role is weighted toward platform and modelling work, and we expect you to move steadily into pipeline and transformation work over your first year or two rather than staying in reporting.
Desirable Criteria
- Microsoft Fabric or an equivalent lakehouse platform - any exposure to Lakehouses, Warehouses or pipelines, including from Synapse, Databricks or adjacent work.
- Agentic / MCP-based tooling - any prior exposure to Model Context Protocol or similar agentic tooling for data work; still rare, so a genuine plus rather than an expectation.
- Version control and deployment practice - any experience of Git, branching or CI/CD applied to data or reporting work.
- Salesforce or Microsoft Dynamics - understanding of CRM data structures helps, given our current landscape.
- Charity or federated organisation experience - comfort with mission-driven reporting, impact metrics and working across a dispersed network.
We bring people together to rescue, repurpose, and share surplus food that would otherwise go to waste, turning an environmental problem into a social opportunity.
Together with 16 other independent charities, 18,000 volunteers, and thousands of partners, we are the UK's food rescue network, supporting 1.5 million people across the country.
We're about fuller plates, fuller communities, and fuller lives., At Felix, we are committed to promoting equality, diversity, and inclusion in everything we do. We value the unique contributions of every individual and strive to create a respectful, inclusive environment free from discrimination or prejudice. Our commitment extends to all employees, and volunteers, ensuring equal opportunities for everyone, regardless of backgrounds or characteristics.