Tweets, press, and people that show why this team is worth your time.
Coinflow is hiring its Machine Learning Lead — the founding leader of Coinflow's ML function and the single owner of the entire fraud and risk intelligence line of business.
This person is going to be owning this business unit. They're going to be the person who's willing to be chewing glass and getting their hands dirty, and then also giving presentations to executives on the status of our system, how we're thinking about taking on risk exposure, how we're calculating projected losses. Setting high-level strategy, communicating it, and communicating what resources they need to achieve it.
The mandate
- Take Coinflow's fraud detection in-house. Coinflow currently uses a third-party vendor. The ML Lead builds the internal replacement from POC through production, ramping volume off the vendor onto Coinflow's own models
- Own the risk-decisioning strategy — approval rates vs. rejection rates vs. loss exposure. Decide the trade-offs. Present them to the exec team.
- Feature engineering + model development + production deployment for authorization fraud, card-not-present fraud, friendly fraud, chargeback patterns, merchant risk
- Full ML lifecycle ownership — experimentation, evaluation, monitoring, retraining schedules, real-time performance metrics
- Define and track fraud metrics — detection rate, false positive rate, chargeback rate, dispute win rate
- Enrich Coinflow's first-party transaction data with partner signals — orchestrate external fraud/risk vendors and get maximum value from their tooling
- Cross-functional: partner with Engineering, Product, Operations, Legal, and Compliance. Not client-facing, but this role sits at the intersection of tech, business, risk, and legal
- Establish ML + data practices for the whole company. Shape Coinflow's long-term fraud, risk, and ML roadmap
Requirements
- 4-8 years of ML / applied data science / production ML experience
- Direct card-acquiring-side payments fraud experience — the single hardest filter. Not general fraud, not general ML — payments fraud on the acquirer / PSP / PayFac side. Authorization fraud, card-not-present, friendly fraud, chargeback patterns, merchant risk
- Proven full-model-lifecycle track record — has personally taken ML systems from POC through production deployment, monitoring, retraining, and iteration
- Leadership track record — has led at least 3-4 engineers on a real project.
- Strategic + technical + business fluency — can go from setting up SOC-equivalent on-call for fraud, to presenting projected losses to the exec team, to negotiating with Legal and Compliance on downstream policy implications
- Chicago-based or willing to relocate
- Deep passion for fraud specifically — Every candidate has been passionate about 'pure machine learning' and asks what else ML will be used for. That's the wrong question. We need someone intensely passionate about specifically payments fraud
Profile Tier 1 — Card-Acquiring / PSP / PayFac ML Fraud Engineers
- Stripe — fraud ML / Radar team. Direct card-acquiring match
- Checkout.com — Direct match
- Adyen — Direct card-acquiring match
- Braintree (PayPal) — direct acquirer/PSP match
- Marqeta, Rapyd, Nium, Airwallex — payments infrastructure with real fraud ML
- Square / Block — direct card-acquiring fraud ML
- Ramp — payments fraud ML
- Visa, Mastercard fraud ML teams — caveated that bank / network engineers can be "square" but the right ones with real product velocity are strong
Tier 2 — Standalone Payments Fraud Services Sardine Riskified, Signifyd, Kount (Equifax) Forter, Sift Science — Feedzai, Featurespace, ThreatMetrix (LexisNexis) — enterprise fraud ML
Tier 3 — Acquiring Banks / ISO / Merchant Services with ML Fraud Depth Chase Merchant Services, First Data / Fiserv, Global Payments (TSYS), Elavon (US Bank), Worldpay (FIS), Bank of America Merchant Services
Tier 4 — Ex-Founders / Founding ML Engineers of Fintech Fraud Startups Tier 5 — Adjacent Fraud ML with Real Payments Exposure Tier 6 — Crypto / Stablecoin Risk ML
Strong Bonuses
- Experience at an acquirer, ISO, PayFac, or payments infrastructure company
- MLOps pipeline and monitoring at scale — retraining schedules, real-time performance metrics
- Cloud compute scoping for scalable ML workloads
- Card network rules fluency — Visa / MC / Amex / Discover network rules, dispute / chargeback workflows, fraud liability frameworks
- Prior early / sole ML hire at a startup
- Real-time or near-real-time fraud scoring systems
- Stablecoin, crypto, or alternative payment rails experience
- LinkedIn presence