Laurie Voss is head of developer relations at Arize AI and a co-founder of npm, Inc. A web developer since 1996, he writes and speaks about developer tools, software careers and the practical engineering of AI systems.[1][2][3]
Start with Stuff Everybody Knows Except You →
Links
Website Bluesky LinkedIn Arize writing
X · seldo Head of Developer Relations · Arize AI
Career
Voss’s first-person biography dates his web-development work to 1996 and lists founding awe.sm in 2010 and npm, Inc. in 2014. He frames tools and teaching as complementary ways to let more people build websites. A September 2023 essay records leaving Netlify and exploring work in AI; Arize’s October 2026 event page identifies his current developer-relations leadership role.[1][2][4]
Contributions
His recurring Stuff Everybody Knows Except You talk introduces professional web development to recent coding-school graduates. Other talks examine the JavaScript ecosystem, npm’s future and scaling decisions. These materials address the knowledge gap between learning to write code and understanding the systems and habits used by working teams. His personal site curates recordings so newcomers can choose a practical overview before moving into specialized topics.[1]
Ideas
His Arize writing examines agent permissions, evaluation and the tools surrounding language models. The author archive includes an article on harness distillation that distinguishes general scaffolding a model might learn from the application-specific tools, data and users a surrounding system still must handle. His evaluation coverage similarly focuses on measuring accuracy, cost and latency under explicit conditions. This places AI engineering in the context of operating and improving software products, rather than assessing a model only from a convincing demonstration.[3]
Tools, writing & talks
Writing, research & talks
Start here
Stuff Everybody Knows Except Youyoutube.comA substantial introduction to professional web-development practice for new developers.
Begin with Stuff Everybody Knows Except You if you have learned coding basics and want a map of professional web development. It is a long overview, so use it to identify unfamiliar practices to study next.
Speaker[1]- How to Know If Your AI Agent Is Any Good: Evals With Phoenix & Claude — Laurie Voss, Arize AIyoutube.com
Arize's Laurie Voss builds a full eval loop on a Claude financial-analyst agent with Phoenix: tracing, code evals, LLM judges, judge validation, experiments, a Haiku 4.5 to 5.5 upgrade check, and CI gating.
Catalog workOriginal source - From Vibes to Production: Evaluating and Shipping AI Agents That Work 201 — Laurie Voss, Arize AIyoutube.com
Arize's Laurie Voss demonstrates adding tracing to an agent, diagnosing a search failure from traces, and having a coding agent fix it. Discusses grouping failures at scale and regression evals so fixes do not break working behavior.
Catalog workOriginal source - From Vibes to Production: Evaluating and Shipping AI Agents That Work 101 — Laurie Voss, Arize AIyoutube.com
Arize's Laurie Voss builds a Claude Agent SDK financial analyst, traces it with OpenInference, and shows why a context-blind judge fails everything. Covers deterministic checks, single-dimension rubrics, human annotations and regression evals.
Catalog workOriginal source Laurie Voss surveys data on code review under coding agents: output up 741%, shipped software up 30%, review quality collapsing on huge PRs, and how Cursor and GitHub run automated review.
Catalog workOriginal source- How long can your skills be before your agent forgets what you told it? — Laurie Voss, Arize AIyoutube.com
Laurie Voss reruns the IFScale instruction-following benchmark a year later and finds frontier models blew past its 200-300 word ceiling, now holding 2,000-5,000 instructions, though new failure modes replace the old one.
Catalog workOriginal source - MCP vs. ACP vs. A2Ayoutube.com
A short overview of communication protocols used around AI agents.
Speaker[1] - Arize articles by Laurie Vossarize.com
Technical writing on agent infrastructure, evaluation and software-development practice.
Author[3]