Staff Software Engineer - Protect

Plaid•San Francisco, CA

About The Position

Plaid Protect is a real-time fraud intelligence product built on a unique advantage: Plaid’s network-level visibility across bank accounts, devices, identities, sessions, institutions, applications, and financial behavior. Protect helps customers detect first-party fraud, synthetic identities, account takeovers, and coordinated attacks that are difficult to see from a single application, account, or transaction. Trust Index turns that fraud intelligence into real-time fraud scores and actionable attributes. This team builds the systems that make this intelligence possible: low-latency inference, new data and model integrations, customer-facing APIs and attributes, safe rollouts, and feedback loops. Ti3 expanded Plaid’s fraud graph nearly 10x and, in early testing, detected up to 41% more fraud at the same false-positive rate. Learn more about Ti2 and Ti3. We are a small, high-agency team working closely with Product, Data Science, and Machine Learning. We value demos over docs, conviction over consensus/alignment, builder schedule over meeting-heavy calendars. We’re scrappy and a talent-dense team that has high agency and high ownership. As a Staff Software Engineer on the Protect Core team, you will set the technical direction for the systems that serve Trust Index scores and fraud attributes in real time. You will lead ambiguous, multi-quarter initiatives while staying hands-on in architecture, critical production code, data investigation, and production debugging. You will work closely with DSML teams, product, and fraud researchers to identify opportunities for new fraud signals and product expansions. You will shorten the path from a promising fraud insight to measurable customer value. You will iterate on Protect, from API changes to new integration modes and product modules. This role is best suited to a hands-on builder who enjoys fast feedback loops, direct debate, and making principled tradeoffs between rapid experimentation and durable systems. This role is not for engineers who prefer long-lead design specs, consensus-driven committee decisions, or 'process-heavy' environments. We prioritize shipping over alignment, and we expect you to push back when the process gets in the way of building.

Requirements

Extensive experience, typically 8+ years, building and operating backend or distributed systems at scale.
Must be comfortable working hands-on directly in the codebase. This is a builder archetype position; the individual will spend considerable time designing and building the actual system they work on.
Strong product judgment: the ability to translate ambiguous customer problems into deliverables, make principled tradeoffs between rapid validation and durable architecture, and define measurable outcomes.

Nice To Haves

Strong opinions, conviction, and good product taste.
Previous experience building data and/or API products.
Prior experience in the fraud domain
Prior experience in or interest in building agentic systems.

Responsibilities

Care deeply about the product and the value you create for our customers.
Learn how customers use our product and the specific fraud challenges they face; build solutions to address them.
Identify new frontiers in fraud detection, build focused prototypes to test the riskiest assumptions, and turn validated ideas into reliable production systems.
Collaborate closely with the DS/ML team. Better yet, navigate our data landscape, run Jupyter notebooks and Spark jobs, or develop in Tecton to deliver business impact independently.
Lead the team by example on AI assisted engineering: write skills and linters, automate things, push code in unfamiliar languages, systems and repos.
Raise the technical bar through mentorship, design reviews, and hands-on leadership across the organization.