Building OpenCode with Dax Raad

2026-05-20 · Source: The Pragmatic Engineer · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Entrepreneurship & Start-ups · Depth: Advanced, extended

Summary

OpenCode co-founder Dax Raad discusses the rapid growth of their open-source coding harness, which has reached nearly 10 million active users in less than a year. Despite this success, Raad highlights that AI coding tools do not inherently accelerate high-quality software development. He notes that OpenCode, like many others, is bottlenecked by GPU supply, emphasizing that AI inference is a highly profitable business with 80-90% margins. Raad also shared an internal memo admitting the team was shipping too many features and hacks, leading to a perceived but not actual increase in speed. He stresses the enduring importance of engineering judgment, strategic thinking, and direct feedback loops in product development.

Key takeaway

For engineering leaders overseeing AI integration, recognize that AI coding tools, while making tasks easier, do not inherently accelerate high-quality software delivery. Prioritize strategic thinking, robust guardrails, and direct feedback loops to counteract the "muted prickle" of AI-driven hacks and prevent feature bloat. Re-evaluate your team's true productivity gains and invest in foundational quality to avoid long-term technical debt.

Key insights

AI coding tools don't inherently accelerate software development; engineering judgment and strategic thinking remain critical.

Principles

• DevTools thrive via bottom-up adoption, not top-down enterprise sales.
• AI mutes the "prickle" of bad code, distorting engineering judgment.
• Combining deep industry expertise with software engineering creates unique value.

Method

OpenCode's product strategy focuses on minimizing user friction, building in public, and leveraging direct user feedback to refine features after initial release.

In practice

• Regularly test first-time user experience with fresh installs.
• Utilize AI agents to efficiently clean up technical debt and refactor code.
• Encourage engineers to cultivate deep industry knowledge beyond coding.

Topics

• OpenCode
• AI Coding Tools
• Software Engineering Productivity
• DevTools
• GPU Supply Chain
• AI Inference
• Technical Debt

Best for: Investor, CTO, VP of Engineering/Data, Director of AI/ML, AI Engineer, Entrepreneur

Related on AIssential

• AI Coding Assistants Haven’t Sped up Delivery Because Coding Was Never the Bottleneck — AI coding tools shift bottlenecks from coding to specification and verification, demanding new team structures and workflows.
• I Spent Two Years Building a Template That AI Replaced in Seconds — Over-abstraction and static templates hinder productivity in dynamic software development, even with AI tools.
• Coding Agents Meet Data Science — AI coding agents boost developer productivity but introduce new bottlenecks and require specialized adaptation for data science workflows.
• the creator of Claude Code just revealed the truth — Strategic talent acquisition and rapid AI advancements are reshaping the tech landscape, from chips to robotics and software development.
• How Ramp engineers accelerate code review with Codex — AI tools like Codex can profoundly enhance developer productivity and code quality by automating complex tasks.
• Boost Productivity with Vibe Coding and IBM Bob — AI coding tools boost productivity for greenfield, low-complexity tasks, but require careful integration and human oversight.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Pragmatic Engineer.