Software Engineer, Agentic Workflows

You'll work alongside engineers, designers, and researchers to design, build, and ship software end-to-end from early prototypes to production systems running in the cloud. The team is investing heavily in agentic systems, the developer experience around them, and the evaluation infrastructure that tells us whether what we're shipping actually works. The work spans agent design, evaluation, cloud services, and developer tooling. You'll have meaningful autonomy in how you get there, and the opportunity to shape both what we build and how the broader org adopts it. The problems are hard, and the team is set up to help you grow into them. Minimum Qualifications 3+ years of software engineering experience with strong CS fundamentals Proficiency in Swift, Objective-C, Python, Go, or another modern language, strong engineers in adjacent stacks will pick up the rest You've shipped software that people used, and you're ready to own bigger pieces end-to-end Strong working proficiency with generative AI coding tools, you've integrated tools like Claude, Cursor, or Codex into how you work, and have a point of view on where they help and where they don't Genuine interest in software evaluation and quality, you want measurement and data to be central to how you ship, not an afterthought Comfortable with ambiguity; when you're stuck, you dig in Strong communication and a track record of working well across teams You seek out feedback and learn fast from the people around you BS in Computer Science or equivalent experience Preferred Qualifications Experience building and deploying cloud-based applications and services at scale Background building developer tools and agentic workflows Familiarity with modern cloud infrastructure (e.g., Kubernetes) Proficiency with one or more scripting languages (Python, Ruby, Bash) Deep expertise with generative AI coding tools, you've pushed past everyday usage into custom agents, evals, or workflows You follow the frontier closely, curious about new models and techniques, with a point of view on where human-AI interaction is headed