Huzzle.com
FP&A Expert (AI Model Evaluation, £70–90/h)
How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Contract · Fully remote · £70–90/h
ROLE OVERVIEW
- Huzzle is seeking FP&A professionals to test how well frontier AI systems handle real financial planning and analysis work, and to document where they fall short.
- You will not be building forecasts for anyone. You will be finding the point at which an AI system stops being reliable at it, and proving that point objectively.
- Each submission has three parts: a realistic financial request, documented evidence of where the model failed it, and a scoring rubric that lets any reviewer grade a future attempt consistently.
- Work spans two reporting bases — US GAAP-based and IFRS-based. Experts fluent in either are encouraged to apply.
- Full training and worked examples are provided. Prior AI or data-annotation experience is not required.
KEY RESPONSIBILITIES
- Design realistic FP&A tasks: driver-based forecasts, budget-versus-actual variance and bridge analysis, scenario and sensitivity modelling, cost and headcount planning, unit economics, capital allocation cases, board-ready reporting packs.
- Ensure each task requires live research against primary sources — filings, published pricing, published rate and index data, regulatory and standards documentation — plus multi-step calculation and an actual file deliverable. Typically a multi-tab Excel workbook, but also memos, decks, PDFs and CSVs.
- Test each task against a frontier AI model and examine its output with the scrutiny you would apply before it went to your CFO.
- Document exactly where and how it failed: figures asserted that do not appear in the cited source, summary cells hardcoded instead of referencing the model, growth applied linearly where it should compound, currencies or reporting bases mixed within a column, variance bridges that do not reconcile, requirements silently dropped.
- Author scoring rubrics with sourced, objective criteria that another analyst could apply without repeating your research.
- Escalate task difficulty where the model succeeds, until a genuine gap is exposed.
Reasons to use Rodeo
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
It searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
EXAMPLE OF THE WORK
A task you might build: "Build a three-year headcount and payroll forecast for a 250-person company growing 12% a year, split by department, with a summary tab comparing two hiring scenarios. Summary figures must be formulas, not typed values."
What the model got wrong: applied 12% growth linearly instead of compounding it, understating year three; typed the summary totals as constants, so changing an input changed nothing; produced four tabs when six were specified.
A line from your rubric: "+3 — Every figure on the summary tab is a formula referencing another tab, not a typed value."
Get help with your application
Your very own career expert that helps elevate your application to the next level.
IDEAL QUALIFICATIONS
- 3+ years in FP&A, corporate finance, financial control or commercial finance.
- Direct experience owning a forecast model, a budgeting cycle or a reporting pack, not only contributing to one.
- Advanced Excel modelling: multi-tab, driver-based, scenario- and sensitivity-enabled. Not template completion.
- The instinct to verify a figure against its source, and to notice when a number is subtly wrong rather than obviously wrong.
- Excellent written English and unusual precision with detail.
- 15–25 hours per week availability.
- ACA, ACCA, CIMA or CFA is a plus, not a requirement. Demonstrated depth matters more than credentials.
CONTRACT TERMS
- You will be engaged as an independent contractor.
- Fully remote, completed on your own schedule.
- The engagement may be extended, shortened or concluded early depending on programme needs and performance.
ABOUT HUZZLE
Huzzle partners with leading AI organisations to evaluate and improve frontier models using deep human expertise. Contributors work directly on the assessment of advanced AI systems in their own field, are paid competitively, and help define the standards by which those systems are measured.
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London
Skills
Location