Skip to content
ClaudeClaude

Building AI-native at enterprise scale: monday.com, Doctolib, and Delivery Hero

Three of Europe's fastest-scaling tech companies made three different bets on Claude: Delivery Hero built an autonomous agent that now merges 100+ PRs a day, Doctolib governs Claude Code across its entire healthcare engineering org, and monday.com ships it inside the product to users who've never written code. Hear what they built, where it broke, and how they stay ahead of a model that changes every few months.

May 20, 202629mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Enterprise leaders share practical playbooks for AI-native transformation with Claude

  1. Delivery Hero built HeroGen, an autonomous software delivery agent that turns Jira/GitHub issues into production-ready pull requests and is now generating high daily merged PR volume with an explicit success-rate metric.
  2. Doctolib focused on company-wide enablement by creating a “skills marketplace,” standardized developer environment, and active internal community to scale effective Claude usage beyond engineering into product and operations.
  3. monday.com used an early “open platform” and exposed APIs to ship monday Vibe, a prompt-to-PRD-to-working-app builder, while confronting new identity and permission challenges when agents become first-class users.
  4. All three teams described model upgrades as non-trivial migrations that require end-to-end evaluation, prompt/workflow re-optimization, and (where possible) A/B testing to avoid regressions.
  5. Across organizations, engineering work is shifting from manual implementation to orchestrating specialist agents, tightening feedback loops (CI/security), and investing in shared context layers that improve agent performance.

IDEAS WORTH REMEMBERING

5 ideas

Integrate agents into existing workflows to maximize adoption.

Delivery Hero kept the primary interface as Jira/GitHub assignment rather than forcing a new chat-first UI, reducing behavior change and speeding uptake across fragmented toolchains.

Define success metrics that reflect real acceptance, not just output volume.

HeroGen tracks merges vs. active rejections to measure whether PRs are truly useful; this creates a tighter quality loop than counting generated code alone.

Use a multi-agent review pattern to reduce blind spots.

Delivery Hero’s “council of agents” has multiple models review the same changes to catch issues the generating model might miss, improving success rates without a prohibitive cost increase.

Standardization and opinionated patterns make agents more effective in complex codebases.

Doctolib observed faster, higher-quality adoption in smaller, consistent services than in a decade-old monolith, where multiple historical patterns confuse model behavior unless explicitly guided.

API-first design becomes a multiplier when building AI-native products.

monday.com’s earlier investment in an open platform (public APIs/SDKs/deployment) let monday Vibe bootstrap quickly by behaving like an external developer, even while the core codebase remained legacy-heavy.

WORDS WORTH SAVING

5 quotes

We looked at where the trajectory is going, and we decided to try out to, uh, build basically an agent that takes, uh, a Jira ticket or a GitHub issue and takes it to, um, uh, production readiness in terms of a pull request that can then be merged, right?

Ulrich Schäfer

One thing that drove that success rate to up to eighty-five percent was, uh, what we call a council of agents.

Ulrich Schäfer

The goal has been really to try to go down the learning curve together rather than everybody kind of doing things and, and doing amazing things, but by themselves.

Alex (Doctolib)

You cannot assume it's just compatible, uh, with the old one. It's literally treating it as a completely different thing, um, and, um, harnessing it and like in a, in a way that works for it, right?

Ruslan (monday.com)

I would say stop waiting for perfect, uh, conditions, perfect use cases, also enough AI-ready work.

Ruslan (monday.com)

Autonomous software delivery (HeroGen)Prompt-to-application building (monday Vibe)Organization-wide Claude adoption and enablementSkills marketplace and standardized dev environmentsModel upgrade rollout, evals, and A/B testingAPI-first platforms, service boundaries, and monolith constraintsAgent identity, permissions, and governance

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.