Arva AI
Data Associate (AI Labelling & Evaluation)
How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Full-time, on-site, London, UK, Aug 5, 2026.
Location: In person, Central London, 4–5 days in office.
Type: Full-Time.
NB: We are able to sponsor visas.
Arva AI is revolutionising financial crime intelligence with our cutting-edge AI Agents. By automating manual human review tasks, we enhance operational efficiency and help financial institutions handle AML reviews, while cutting operational costs by 80%.
As our first dedicated Data Associate, you'll build the ground truth our AI Agents are trained and measured against — labelling real financial crime casework, grading agent decisions, and turning tricky edge cases into evaluation sets. Every label you create makes our agents measurably better, safer, and easier to audit. You'll be operating at the intersection of compliance operations and applied AI in a fast-moving early-stage environment.
About the Role
As a Data Associate, you will:
- Own the labelling and annotation of financial crime data — KYB/KYC cases, screening hits, and transaction alerts — that powers the training and evaluation of our AI Agents.
- Become our internal source of ground truth — defining what "correct" looks like for agent decisions and holding every model release to that standard.
- Partner closely with Engineering, Product, and Compliance to turn real-world casework into structured datasets that measurably improve agent performance.
What You'll Do
- Labelling & Annotation: Review and label KYB/KYC cases, sanctions, PEP and adverse media screening hits, and transaction monitoring alerts to create high-quality training and evaluation data.
- Agent Evaluation & QA: Grade AI Agent outputs against gold-standard answers, flag errors and inconsistencies, and track quality metrics across model releases.
- Guidelines & Taxonomy: Help define and refine labelling guidelines, decision taxonomies, and edge-case playbooks so that labels stay consistent as the team and dataset scale.
- Edge Case Discovery: Surface ambiguous, novel, or adversarial cases from live data and turn them into structured evaluation sets that stress-test our agents.
- Feedback Loop: Communicate patterns of agent errors to Engineering and Product, and help prioritise the data work that will most improve performance.
- Data Integrity: Keep datasets clean, versioned, and well-documented — ensuring auditability and handling sensitive customer data with rigour and care.
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.
Our Culture
- Deliver Value Fast: Speed starts with clarity. We first understand what value actually means, for the customer, the business, or the system, and then take the shortest credible path to delivering it.
- Outcome Obsessed: We obsess over details and take full ownership of wider outcomes, not just tasks. If something falls short, we fix it properly and prevent recurrence, always raising the bar.
- Relentless Urgency: We move with urgency because time matters. We prioritise what truly moves the outcome, make decisions with imperfect information, and act decisively.
What We're Looking For
- 1- 2 years of experience: 1-2 years of professional experience in the workplace.
- High-rigour early-career candidate: A strong recent graduate or early-career candidate (e.g., law, finance, criminology, data, linguistics) with demonstrable attention to detail and a genuine interest in AI. No compliance experience required — but you'll need to learn fast.
- Nice to have: Understanding or experience as a KYC/KYB, AML, or fraud analyst at a bank, fintech, or compliance vendor.
- Rigour: Obsessive attention to detail, and the ability to stay sharp and consistent through high volumes of consequential review work.
- Judgement: Comfortable making defensible calls on ambiguous cases — and documenting your reasoning clearly so others can follow it.
- Curiosity: Genuine interest in financial crime, compliance, and how AI systems learn — you don't need to be an AML expert on day one, but you should be excited to become one.
- Ownership: Proactive, self-directed, and accountable for outcomes. You don't wait to be told what to do.
- Communication: Clear written documentation and the ability to articulate precisely why a label or verdict was chosen.
- Tooling: Comfortable working in spreadsheets and annotation tools; SQL or Python is a plus, not a requirement.
Get help with your application
Your very own career expert that helps elevate your application to the next level.
Why Join Us?
- Be part of an early-stage startup with significant ownership and direct influence over how our AI Agents learn and improve.
- Work on a product that directly impacts how financial crime is detected and prevented globally.
- Collaborate with a passionate, mission-driven team operating at the intersection of AI, compliance, and enterprise software.
- Work from anywhere in the world for 4 weeks a year, in addition to regular team off-sites.
- Competitive salary and equity package, with bi-annual salary review and yearly performance-based equity refresh.
Ready to Join the Fight Against Financial Crime? If you're excited to build the ground truth behind AI Agents fighting financial crime, we'd love to hear from you. Apply now to become an Arvanaut as our Data Associate.
“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