# How to build an AI-Native Health Company — Dan Feng, Maven Clinic

Source: [AI Engineer](https://www.youtube.com/watch?v=WJRdLNhrsLQ)  
Feed7 permalink: https://feed7.dev/p/how-to-build-an-ai-native-health-company-dan-feng-maven-clinic-0hr5j7m  
Published: 2026-08-19T17:30:19.000Z  
Trust: Source Linked (source_linked)

## Why Included

Maven’s agent-heavy workflow shortens planning, caps reviewable PR size, and tests nondeterministic features repeatedly. The bottleneck shifts from implementation to decisions and verification.

## Source Summary

Maven describes replacing long planning cycles with **2–4-week delivery windows**, **one-to-two-page specs**, and reviewable PRs capped at **500 lines**. Its LLM integration tests target a sustained **90% pass rate**, not a single green run.

## Practical Implication

Keep agent-generated changes small, make short-term requirements concrete, and spend engineering attention on architecture, evaluation, and consequential failure paths. Low-risk changes may justify self-review, but ownership stays with the author.

## Agent-Ready Context

Maven describes replacing long planning cycles with **2–4-week delivery windows**, **one-to-two-page specs**, and reviewable PRs capped at **500 lines**. Its LLM integration tests target a sustained **90% pass rate**, not a single green run.

Keep agent-generated changes small, make short-term requirements concrete, and spend engineering attention on architecture, evaluation, and consequential failure paths. Low-risk changes may justify self-review, but ownership stays with the author.

These are one healthcare company’s operating rules, not established defaults. Maven still values human review, says automated review is not yet sufficient, and has not completed its goal of automating the full delivery lifecycle.

## Connected Context

Feed7 judgment across 525 accumulated Signals:

This turns bounded agent engineering into specific operating constraints: short delivery windows, compact specs, small PRs, and repeated integration-test performance. It confirms that higher automation does not remove authorship or human review; engineering effort shifts toward architecture, evaluation, and costly failure paths. The numbers are local practices, not general thresholds.

- [Loop Engineering from First Principles — Kyle Mistele, HumanLayer](https://feed7.dev/p/loop-engineering-from-first-principles-kyle-mistele-humanlayer-1nuq7gf) — Maven’s small PRs and short, concrete requirements operationalize the bounded-task and human-pause loop proposed by HumanLayer.
- [How to Kill the Code Review — Ankit Jain, Aviator](https://feed7.dev/p/how-to-kill-the-code-review-ankit-jain-aviator-0rku6kj) — Both redirect review toward intent and verification evidence, but Maven retains human review and author ownership rather than treating automated review as sufficient.
- [Don't Build Agents You Can't Answer For — Addy Osmani](https://feed7.dev/p/don-t-build-agents-you-can-t-answer-for-addy-osmani-1y9nwej) — The sustained pass-rate target and ownership rule provide concrete forms of the tests, evidence, and accountable authorship Osmani says must accompany agent throughput.
- [Harness Engineering is not Enough: Why Software Factories Fail — Dex Horthy, HumanLayer](https://feed7.dev/p/harness-engineering-is-not-enough-why-software-factories-fail-dex-horthy-1tj1amt) — Maven reinforces the maintainability warning by keeping changes reviewable and humans accountable even while pursuing broader lifecycle automation.

## Context Map

- Layer: agent
- Domains: coding
- Topics: coding-agents, harness-engineering, agent-reliability

## Uncertainty

- These are one healthcare company’s operating rules, not established defaults. Maven still values human review, says automated review is not yet sufficient, and has not completed its goal of automating the full delivery lifecycle.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.