Startups

Senior Software Engineer, Agentic Applications

Allen Institute for Artificial Intelligence · Seattle, WA · On-site

← All jobs
About Allen Institute for Artificial Intelligence

Ai2, founded by Paul Allen, conducts high-impact research and engineering to tackle key problems in artificial intelligence.

About the role

The Allen Institute for AI (Ai2) is hiring a Senior Engineer to build AI-powered products that advance scientific discovery. You'll own end-to-end product surfaces: designing interactions for long-running, non-deterministic agents, contributing to the agent architecture behind them, building React applications on top of our agent infrastructure, and shaping how users understand and steer what the system is doing. You'll work closely with UX designers, AI/ML engineers, and research scientists.

What they're looking for

  • Bachelor's degree and 8+ years of technical experience, relevant experience may substitute for education
  • Professional experience with web development using JavaScript, HTML, and CSS
  • Strong Python engineering skills
  • Proficiency with front-end frameworks such as React
  • Track record shipping user-facing products, with a demonstrated eye for interaction design
  • Experience building and operating agentic products in production
More about this role

The Allen Institute for AI (Ai2) is hiring a Senior Engineer to build AI-powered products that advance scientific discovery. You'll own end-to-end product surfaces: designing interactions for long-running, non-deterministic agents, contributing to the agent architecture behind them, building React applications on top of our agent infrastructure, and shaping how users understand and steer what the system is doing. You'll work closely with UX designers, AI/ML engineers, and research scientists. You're a seasoned engineer who thrives in greenfield environments and has a track record of building intuitive, high-performance user experiences. You're energized by the design problems that come with agentic products: streaming state, ambiguous outputs, and users who need to trust the work.

The Agentic Applications team builds Asta , Ai2's open research assistant for science, which finds, summarizes, and analyzes scientific evidence across millions of papers. Go try it; it's public. Our hardest problems are interface problems: a scientist won't trust an answer they can't verify, so our agents have to show their work — and real research doesn't finish in a chat session, so they have to run...

Read the full posting on Allen Institute for Artificial Intelligence's site ↗

Engineering

Build your edge while you search

Free tools for founders and investors, plus VC Unfiltered, our take on startups, venture and the people who build them.