We're looking for an experienced AI/ML engineer with deep domain expertise in financial crime to join our fintech startup at the ground floor. If you've built fraud models before and want to apply that expertise to the biggest unsolved problem in payments, read on to learn more ⬇️
About the role 🎭
Banks approve or reject payments with almost no context. We're building the intelligence layer that changes that — running real-time investigations on payments before they clear. You'll own the models and machine learning systems that power our fraud prediction: turning messy, multi-source evidence into risk scores that banks trust and act on.
What you'll work on
Design, build, and iterate on ML models and AI products that underpin our fraud detection. Feature engineering, guardrails, evals, production deployment, monitoring — you’ll work on the full end-to-endDevelop fraud prediction capabilities that go beyond sender-side signals: scoring recipient risk, payment context, and situational indicators using labelled fraud data and feedback loopsImprove the capabilities of our investigation engine to turn unstructured, multi-source evidence into features that improve model performanceBuild and maintain the MLOps infrastructure needed to train, evaluate, deploy, and monitor models in production with the reliability banks demandCollaborate closely with engineers, product, and our bank customers to understand where model improvements have the highest impact on fraud prevention outcomesBring your domain expertise to bear: help shape our product roadmap by identifying which fraud typologies, signals, and intervention strategies will move the needle mostExperiment with and deploy LLM-powered approaches where they outperform traditional ML or provide faster time-to-market — we're pragmatic about techniques, not dogmaticContribute to our data strategy: what labels do we need, how do we close feedback loops, and how do we build compounding data advantages over timeThis is not a research role. You'll ship models that make real-time decisions on real payments. Engineers here drive product decisions and own systems end-to-end.
Why this is hard
Fraud is adversarial, low-prevalence, and high-stakes. Labels are noisy and delayed. The evidence that distinguishes a scam from a legitimate payment is messy — unstructured, scattered across sources, rarely behind clean APIs (but you know this already!). Models need to be fast enough for real-time payment flows, explainable enough for banks to trust, and robust enough that adversaries can't trivially evade them. And we're critical infrastructure: when we're wrong, real people lose money.
Our stack today:
Backend: Python, KotlinInfra: Cloud-native, event-driven services, async workflows (GCP managed using Terraform)Data: BigQuery, Elastic; correctness > dashboardsML: Early and evolving — you'll have a major hand in shaping thisIf you've built production fraud models before, the specific tools matter less than your judgement on what works.
About YOU 🦄
Here are some thoughts on who would be successful in the role. We know we won't be able to assess these from your CV, but our interview process will be designed around them:
Domain conviction: You've seen how fraud works up close — APP scams, payment fraud, financial crime — and you have strong opinions on what signals matter, what interventions work, and where the industry is getting it wrong. You want to apply that knowledge at a company where it will shape the product, not just improve a metric.Product ownership: You ask "what fraud outcome does this model improve?" before "what's the AUC?" You care about impact on real users, not just model performance in isolation.Startup energy: You're excited by ambiguity, speed, and building from scratch. You've ideally done this before at an early-stage company and know how to make pragmatic trade-offs between rigour and velocity.Collaboration: You're direct, proactive, generous to others, and low ego. You'll work across engineering, product, and customer teams — and you'll need to translate between ML-speak and business outcomes fluently.Some hard skills we'd value
Fraud/financial crime ML: Production experience building fraud detection or financial crime models — ideally APP fraud, payment fraud, or transaction monitoring. You understand the specific challenges: class imbalance, adversarial drift, noisy labels, delayed feedback.Feature engineering: Strong instinct for what makes a good feature in fraud contexts. Experience working with heterogeneous data sources (structured transactions, unstructured text, network/graph signals, third-party enrichment).MLOps: You've deployed and monitored models in production. You know how to detect drift, manage retraining pipelines, and ensure model reliability at scale.Classical ML + modern approaches: Solid foundations in gradient-boosted models, logistic regression, and the bread-and-butter of fraud ML. Bonus if you've also shipped LLM-powered or deep learning approaches in production.Data engineering fluency: Comfortable with SQL, BigQuery, and building the data pipelines that feed your models. You don't need a separate team to get data into shape.Years of experience: 7 or more. We're looking for someone who brings genuine subject matter expertise — someone who has seen enough fraud patterns, built enough models, and shipped enough systems to have strong, informed opinions (we're not overly strict on exact number of years!).Stage of experience: We strongly value people who have worked in early-stage companies or have demonstrated the ability to build from zero. You're comfortable owning the full stack from data to deployment.Industry: Prior experience at a fintech, neobank, or payments company building fraud/financial crime systems is a huge plus.Not got 100% of qualifications? Research shows that women and under-represented minorities are less likely to apply for a role if they do not have 100% of the requirements. We're paying attention to this, so please feel free to apply regardless (<2mins) or send us a note on info@tunicpay.com if you'd like to discuss if there's a fit.
