Platform Engineer

Why This Role

Concourse is building the operating system for State and Local governments. Our platform is a configurable approach to tackling fragmented government data and workflows and turning them into smooth operations for public servants and the constituents they serve. As a Concourse Platform Engineer, you're not building a marginal feature for one product. You're contributing to the infrastructure that scales to our hundreds of products serving millions of Americans. You own the foundational layer: the data ingestion pipelines, AI workflow orchestration engine, and configurable primitives that support everything above them. Your role is to empower our Forward-Deployed PMs who sit alongside city and county teams, understand their specific workflows, and assemble each agency's solution. A bail management system for a county courthouse. IT Helpdesk for a school district. A water utility operations dashboard. An ERP for a mid-sized city. The FDPMs build those. Your job is to make sure they can. You're working on the deep infrastructure questions that make every deployment possible: how to reliably ingest data from a 30-year-old mainframe, how to run multi-step AI workflows that government users can audit and trust, how to expose primitives flexible enough to model any government process but constrained enough that non-engineers can build on them. These are hard, novel problems. The people solving them will define the next generation of government software serving us.

What You'll Do

• Design and own the connectors, ingestion pipelines, and data modeling primitives that normalize government data from legacy systems (mainframes, Oracle, flat files, proprietary APIs) into a consistent, queryable foundation
• Build the infrastructure for long-running, multi-step AI workflows: task queuing, state management, retries, human-in-the-loop checkpoints, and the audit trails government compliance requires
• Build and maintain the core primitives FDPMs build on top of: the workflow engine, role-based access model, form and action framework, and event system that let a non-engineer assemble a bail system or permitting platform without writing the plumbing from scratch
• Bake in data isolation, access controls, and auditability as core platform features from the start, with a path toward compliance certifications (FedRAMP, CJIS, HIPAA) as we grow
• Work directly with the founder on architecture decisions that will compound across every deployment we ever ship

What We're Looking For

• A portfolio: what have you built end-to-end, from scratch, under pressure?
• Strong backend fluency in Python and/or Node, with solid instincts for data modeling, API design, and building systems that stay reliable and extensible over time
• Enough frontend ability to build and reason about a UI, even if you won't own it
• Hands-on experience using AI coding tools (Claude Code, Codex, Gemini CLI, Factory, etc.) as a core part of your daily workflow
• Good instincts for abstraction: knowing when to generalize something into a platform primitive vs. when to just ship it
• US citizen (required for government work)
• Willing to work in-person in Flatiron, NYC (5 days/week)

Nice to have:

• Experience building developer platforms or internal tooling, devops/infrastructure, agentic AI orchestration, or work in regulated industries like healthcare, finance, or legal.

About Concourse

Concourse is the fastest-growing govtech, providing software to more than 100 state and local governments across the U.S. Government calls what we do "IT modernization." We call it a trillion-dollar opportunity to overhaul how government works. Public services are broken not because of bureaucracy but instead because they run on antiquated, user-unfriendly software from legacy vendors. Some examples: Concourse is powering the legislative management system for one of America's largest cities, a state's entire online bail payment system, and a case-management platform for mental-health response teams. We are an AI-native company. Our org structure and staffing model are radically different from those of legacy tech companies: we have a flat team of superstar generalists who both own client relationships and manage fleets of long-horizon AI agents. Our team works together in our Flatiron office five days a week.