Delivery Hero
Code review
Manual code reviews missed bugs and slowed delivery. AI now drafts pull request summaries and flags missing tests across 15,000 repos.
- AI enabled on 15,000+ GitHub repositories
- 4,000+ developers using AI assistant
Refactoring hundreds of decentralized codebases took days. Now, an AI agent writes and submits code in minutes for human review.
A global logistics network operating in 65 countries processing over 11 million daily orders, with thousands of engineers spread across dozens of decentralized regional operations using disparate cloud providers and programming languages.
Centralizing the developer platform caused friction among teams with deeply held preferences for different build systems and infrastructures....
“There's always a certain history and attachment to whatever engineers have at the moment. Convincing everybody that a global developer platform might be a good idea took around a year and a half.”
Global food and grocery delivery platform with quick commerce and Dmarts.
Anthropic is a technology company specializing in artificial intelligence and machine learning solutions.
Delivery Hero's Code generation is part of this use case:
Related implementations across industries and use cases
Manual code reviews missed bugs and slowed delivery. AI now drafts pull request summaries and flags missing tests across 15,000 repos.
Scale buried the CTO in review overhead—no production code since the dorm room. Claude Code brought him—and DoorDash—back into production.
Engineers maintained 40,000 microservices by hand, slowing new features. Now Claude handles migrations, reviews, and tests at scale.
Manual code reviews missed bugs and slowed delivery. AI now drafts pull request summaries and flags missing tests across 15,000 repos.
Scale buried the CTO in review overhead—no production code since the dorm room. Claude Code brought him—and DoorDash—back into production.
Routine calls crowded out what Trinity was built on. AI took the volume; the team reclaimed time for the carrier relationships that count.
Accident probes left drivers on leave for weeks. Now any operations manager can query the live fleet in plain language—cases close in days.
Tournaments running simultaneously meant an hour of manual checks each. AI agents now run them in minutes, freeing the team to be proactive.
Large AI training jobs meant fighting for preemptible slots or leaving campus. Marlowe gave any lab guaranteed multi-node access on demand.
Refactoring hundreds of decentralized codebases took days. Now, an AI agent writes and submits code in minutes for human review.
A global logistics network operating in 65 countries processing over 11 million daily orders, with thousands of engineers spread across dozens of decentralized regional operations using disparate cloud providers and programming languages.
Centralizing the developer platform caused friction among teams with deeply held preferences for different build systems and infrastructures....
“There's always a certain history and attachment to whatever engineers have at the moment. Convincing everybody that a global developer platform might be a good idea took around a year and a half.”
Global food and grocery delivery platform with quick commerce and Dmarts.
Anthropic is a technology company specializing in artificial intelligence and machine learning solutions.
Delivery Hero's Code generation is part of this use case:
Related implementations across industries and use cases
Manual code reviews missed bugs and slowed delivery. AI now drafts pull request summaries and flags missing tests across 15,000 repos.
Scale buried the CTO in review overhead—no production code since the dorm room. Claude Code brought him—and DoorDash—back into production.
Engineers maintained 40,000 microservices by hand, slowing new features. Now Claude handles migrations, reviews, and tests at scale.
Manual code reviews missed bugs and slowed delivery. AI now drafts pull request summaries and flags missing tests across 15,000 repos.
Scale buried the CTO in review overhead—no production code since the dorm room. Claude Code brought him—and DoorDash—back into production.
Routine calls crowded out what Trinity was built on. AI took the volume; the team reclaimed time for the carrier relationships that count.
Accident probes left drivers on leave for weeks. Now any operations manager can query the live fleet in plain language—cases close in days.
Tournaments running simultaneously meant an hour of manual checks each. AI agents now run them in minutes, freeing the team to be proactive.
Large AI training jobs meant fighting for preemptible slots or leaving campus. Marlowe gave any lab guaranteed multi-node access on demand.