Senior Full Stack Engineer

Techstars is building an operating system that centralizes how we run accelerator programs, support founders, engage the network, and drive investment decisions. As a Senior Full Stack Engineer you'll own features end to end — from the data behind them to the React UI in front of them, from design to go-live. That includes the AI-native parts of the product: search that spans the network, retrieval over founder and portfolio data, and assistive workflows that put the right context in front of the right person.

You'll also lead the wind-down of our legacy founder tools, consolidating what matters into OS and retiring the rest cleanly.

This is a small, high-trust team. We move fast, ship real things, and expect everyone to be able to hold a thread all the way through.

What you will do

• Own features on Techstars OS end to end — design, build, test, and ship, from data ingestion and API layers through to the web app
• Build AI-native features into the product: semantic and federated search across the network, retrieval over portfolio and founder data, and assistive workflows that surface the right context at the right moment
• Lead the migration and shutdown of legacy founder tools — including Techstars Universe and our application management system — consolidating functionality into OS or retiring it cleanly
• Work across our Enterprise Data Platform and OS to build the pipelines that feed real-time features, search, and AI use cases
• Raise the floor on reliability and velocity: tests, code review, and CI/CD practices that hold up as the team grows — including how we fold AI coding tools into that loop
• Act as a technical resource for the team — unblock others, weigh in on architecture, and keep context moving across time zones

What you bring

• Strong TypeScript and Python, and the judgment to know when to reach for each
• Solid React experience — you can build clean, functional UIs without a dedicated frontend team carrying you
• Proficiency with relational databases: data modeling, migrations, and query performance under real load. We use PostgreSQL
• Experience building data pipelines end to end — ingestion, transformation, and delivery to downstream consumers. We use event streaming, but experience with any comparable system transfers
• Experience shipping LLM-backed features to production: retrieval, embeddings, and the evaluation loop that keeps quality honest. You understand the hard part is uncertainty, evidence, and recovery — not the API call
• Comfortable operating containerized services in a major cloud with infrastructure defined as code. We run on GCP and AWS
• A track record of owning features from zero to production, and of retiring old systems without breaking things downstream
• Strong written communication and a high tolerance for autonomy — remote-first means writing well and staying aligned without being chased

Nice to have

• Event-driven architectures and Kafka-based data platforms
• Evaluation and observability for LLM features — you've had to answer "is this actually working?" with data
• MCP or other tool surfaces that let agents act on production systems safely
• Identity and auth systems (Keycloak or similar)