← back
Beyond the Prototype: Using AI to Write High-Quality Code - Josh Albrecht, Imbue
Takeaway
Quality AI-generated code comes from synchronous prevention plus detection loops — plan, spec, lint, test, LLM-review — not from picking the smartest base model.
Summary
- Imbue's Sculptor is a coding-agent environment built around the question 'what's wrong with this diff?' — adding a third option between hand-reviewing every line and blind-merging.
- Four prevention techniques baked in: research/learning, plan-mode system-prompt switching, spec-and-doc drift detection, and a strict (functional/immutable) AI-tailored style guide.
- Three detection techniques: auto-running linters (with before/after diff so AI can only reduce errors), test running, and LLM-as-reviewer feedback delivered synchronously rather than at PR time.
- Argues teams shouldn't build commodity coding-agent infra (context length, parsing) that frontier models will solve in 3–24 months; build the verification and trust layer specific to your domain.
coding-agentssculptorcode-quality
Original description
In this case study-based keynote, Josh Albrecht, CTO of Imbue, examines the critical engineering challenges in building AI coding systems that create more than just prototypes. Drawing from Imbue's research developing Sculptor, an experimental coding agent environment, Josh shares key insights into the fundamental technical obstacles encountered when evolving AI-assisted coding from toy applications to more robust software systems. The session will explore approaches to core challenges like safely executing code, managing context across large codebases, automating test generation, and creating systems that can identify potential pitfalls in AI-generated code. Attendees will gain practical insights into the technical underpinnings of next-generation coding agents that aim to handle complex software engineering challenges architecting larger systems, increasing meaningful test coverage and designing systems that are easy to debug—moving us closer to AI systems that can help create maintainable software. About Josh Albrecht Josh Albrecht is CTO and Co-founder of Imbue, an AI lab launched in 2022 that has since raised $230M at a $1B valuation to create coding agents that make it easier for more people to write software. Josh is also a partner at angel fund Outset Capital, where he invests in promising pre-seed companies. Previously, Josh founded multiple companies including an AI recruiting startup that went through Y Combinator and a 3D injection molding software company that was acquired. He was also an early engineer at Addepar, served as a Thiel Fellow mentor, and published machine learning research as an academic researcher. Recorded at the AI Engineer World's Fair in San Francisco. Stay up to date on our upcoming events and content by joining our newsletter here: https://www.ai.engineer/newsletter