Software Engineer

As a software engineer, you will :

Build the core simulation platform : Create a realistic, containerized environment where AI agents perform software engineering and other professional tasks.

Design realistic evaluation scenarios : Independently identify meaningful software engineering tasks, determine clear grading criteria, and select the appropriate tools and workflows that allow AI agents to reliably complete tasks.

Develop a consumer software product : This product will serve as a foundation for our simulation platform, where we will task AI agents with building and modifying this product. This will require full stack skills from writing a backend in Python and a frontend with web technologies like React, Typescript, HTML and CSS.

Establish engineering standards : Set up automated testing, practical CI / CD pipelines, effective monitoring, and clear processes to ensure high-quality code and quick issue resolution.

Influence technical direction : Directly shape product development, make key architectural decisions, and contribute to growing our engineering team at an early-stage startup.

You might be a fit if you :

Have 5-10 years of experience building full-stack software products, including leading at least one significant project from idea to production.

Write fluent TypeScript (React) and Python , and are comfortable with deployments in Kubernetes-based environments.

Have practical experience with designing robust CI / CD processes, automated testing strategies, and effective monitoring and observability.

Enjoy working closely with users (internal stakeholders and customers) to iterate quickly and deliver functional software.

Are able to at least solve basic codeforces-type problems (e.g. problems at a 1500 ELO)

Bonus points for experience with :

Container orchestration platforms (Kubernetes)

Developer tools or internal platforms

Reinforcement learning (RL) or machine learning (ML) workflows and infrastructure

We encourage you to apply even if you do not believe you meet every single qualification. Not all strong candidates will meet every single qualification as listed.