Lead a strong team in building real-time risk decision systems for a growing FinTech company.
Posted by employer 3 days ago
First seen on Joblaze 1 day ago
Last verified on the company career page 1 day ago
What you'll build
Must have
Nice to have
Role intensity
10% coding — mostly leadership/strategy
AI in the day-to-day
Managers at Imprint use AI tools to investigate codebases, assess complexity, and clarify ambiguous asks.
Requirements
Not disclosed in this posting: compensation, work arrangement, visa sponsorship.
Benefits
Joblaze summary
The Senior Engineering Manager for Risk Engineering at Imprint oversees the development and operation of critical systems that handle real-time transaction authorization and fraud scoring. This role requires a strong technical background, particularly in managing latency-sensitive distributed systems, and familiarity with tools like Go, MySQL, and AWS. Ideal candidates have extensive engineering management experience, particularly in high-stakes environments, and a proven ability to lead and develop senior engineering teams. The position is situated within a robust team that values psychological safety and technical excellence.
Joblaze insights
Quick facts
From the original posting
Risk Engineering owns the systems that execute Imprint's risk decisions in real time: transaction authorization and fraud scoring on every card swipe, the underwriting pipeline that evaluates every application, and the vendor integrations that power both. These are latency-sensitive, high-stakes distributed systems with a sprawling surface area and many stakeholders competing for the team's capacity. AI has made execution faster; the differentiator is the judgment to decide what to build, how to sequence it, and when to say no. That judgment, applied to a strong team and a hard domain, is what this Senior Engineering Manager owns.
Own real-time authorization and fraud decisioning: Every card swipe runs through systems your team builds. Decisions happen in milliseconds, failures are visible immediately, and a wrong answer costs real money in either direction: fraud losses on one side, declined good customers on the other. You set the bar for how these systems are designed, tested, and operated.
Own the underwriting pipeline: Every application Imprint receives is evaluated by the pipeline your team owns, spanning credit, fraud, identity, and compliance checks. As Imprint launches new brand partners and new products, the pipeline has to absorb new policies and new data sources without becoming brittle.
Make vendor decisions that hold up in production: Fraud scoring, identity verification, and bureau data all run through third-party integrations. You'll evaluate vendor capabilities, weigh build-versus-buy with real latency and cost numbers, and push back on approaches that won't survive production load.
Prioritize across a wide surface cross-functional surface area: Credit, Fraud, Product, and Compliance all have legitimate asks and compete for the same engineering capacity. You'll bring clarity to poorly-scoped platform problems, decide what the team does and does not take on, and make the reasoning legible to stakeholders who didn't get what they asked for. You define the roadmap and scope the work yourself; nothing is abstracted through organizational layers.
Lead with technical credibility, amplified by AI: AI has collapsed the distance between management and IC work. Managers at Imprint use AI tools to investigate codebases, assess the complexity and viability of work, and bring clarity to ambiguous asks. You don't need to be on the critical path writing code (though some managers are), but you need the technical judgment to evaluate design trade-offs on latency-sensitive distributed systems and the modern tools to exercise it.
Lead a strong team as a craft: You inherit a strong, psychologically safe team of senior engineers. The job is to keep it that way while raising the ceiling: hand ICs ownership of entire domains, hold the bar on how work is scoped and delivered, and give feedback that changes outcomes. You have a real philosophy on how to hire, develop, and retain strong engineers, and clear written and verbal communication is how you exercise it.
5+ years of engineering management experience, with a track record of building, developing, and retaining strong engineering teams
You've managed through periods of rapid growth, ambiguity, or significant technical change, and your teams came out stronger
A strong technical foundation: you were a strong engineer before moving into management, and you've maintained technical credibility through hands-on engagement or modern AI tools
You've managed Staff-level ICs and helped senior engineers grow into bigger roles
You've owned latency-sensitive, high-availability distributed systems in production, where the cost of a wrong decision is measured in real dollars
You communicate with clarity and directness in writing, in 1:1s, in rooms with executives, and in rooms with engineers
You drive delivery through your team, not around them. You know how to scope work, set priorities, and create the conditions for engineers to do their best work
Experience using AI tools to stay close to technical work (investigating codebases, assessing feasibility, bringing clarity to asks), or genuine enthusiasm to adopt them
Experience in risk, fraud, underwriting, or payments is a plus
This is a backend team. We don't expect every candidate to match every bullet. If you're excited about what we're building and confident you can contribute, we'd love to talk.
Our backend runs on Go, MySQL, DynamoDB, Temporal, Kafka, Kubernetes, and AWS. Read more about how we build at Imprint on our engineering blog.
Competitive compensation and equity packages
Leading configured work computers of your choice
Flexible paid time off
Standard company text repeated across Imprint's postings is omitted here.