How do 94 engineers actually use coding agents? · Prehistoriq Stacks

Peter Steinberger

OpenClaw, ex-PSPDFKit

OpenClawwebCodexappGPT-6.1 Solworktrees

ReviewsCodex reviews every commit

Simon Willison

Datasette, Django co-creator

CodexappClaude CodewebGPT-6 SolOpus 5.5phone remote

ReviewsCross-model review: Claude checks GPT, GPT checks Claude

Boris Cherny

Created Claude Code

Claude CodeappClaude TagFable 5.1worktrees

ReviewsReads diffs only for critical code

Andrej Karpathy

nanochat, coined “vibe coding”

Claude CodecliCodexFable 5Ghostty + IDE

ReviewsWatches agents like a hawk in an IDE

Mitchell Hashimoto

Ghostty, HashiCorp co-founder

Claude appappCodex appappFableGPT-6 Astradesktop apps

ReviewsHand-reviews every line; agents validate, he writes all prose

DHH

Ruby on Rails, Omarchy

CodexcliClaude CodecliOpenCodecliGPT-6.1 SolOpus 5.5

ReviewsA second model reviews, then he reads the diff

Armin Ronacher

Flask, Earendil founder

PicliCodexSolFableGhostty + tmux

ReviewsSteers as it goes, reviews when the agent says done

Theo Browne

T3 Stack, T3 Chat, T3 Code

Claude CodeappCodexappOpus 5.5Muse Wonderremote Linux

ReviewsOpus asks Wonder for code review; agents merge when confident

Salvatore Sanfilippo (antirez)

Redis creator, DwarfStar (ds4)

CodexcliClaude Codeclids4-agentcliGPT-6 Astra / GPT-6.1 SolFable

ReviewsStill reviews Redis code, calls it mostly pointless

ThePrimeagen

Harpoon, Vim on stream

Cursoride99ideGrok 4.6terminal + 99

Steve Yegge

Gas Town, Beads, Vibe Coding

Claude CodecliCodexOpus 5.5Fable 5.1Emacs + tmux

ReviewsFable reviews every Opus change; he doesn't read code

Pieter Levels

Photo AI, Nomads.com

Claude CodecliOpus 5.5Fable 5.1VPS + tmux

ReviewsStep by step, keeps the agent on a short leash

Mario Zechner

pi coding agent, libGDX

PicliGPT-5.5phone

ReviewsLimits daily output to what he can review

Thorsten Ball

Amp co-founder, interpreter book

AmpwebOpus 5.5GPT-6 Astraorbs

ReviewsSpot checks and architecture, not every line

Jesse Vincent

Superpowers, Keyboardio, K-9 Mail

Claude CodecliEvenerMuseOpusHaiku

ReviewsAgent sessions do adversarial review

Harper Reed

CEO 2389.ai, CTO Obama 2012

Claude CodecliCodexappFable 5.1GPT-6 Astratmux + mosh

ReviewsOne fresh-eyes review pass per change

Kent Beck

TDD, Extreme Programming, JUnit

Claude CodecliCodexcliGeminilive sessions

ReviewsReviews solo, would rather be pairing

Kieran Klaassen

Compound Engineering, Cora

CursorClaude CodecliCodexOpus 5.5Sonnet 5.5

ReviewsParallel review agents, no line-by-line reading

Matt Pocock

Total TypeScript, ex-Vercel

Claude CodecliOpus 5.5Opus 5Sandcastle

ReviewsReviews by merge danger, two-axis /code-review

Indragie Karunaratne

Context, uniprof, Sentry AI lead

Claude CodecliCodexOpus 5.5GPT-5GitHub Desktop

ReviewsReviews the plan and prompts, not the 10,000-line diff

Addy Osmani

Claude Code at Anthropic, ex-Chrome

Claude CodeCodexFableworktrees

ReviewsCareful human review of core paths only

Geoffrey Huntley

Creator of the Ralph loop

OpenCodecliCursorwebLoomGPT-6.1 SolGLM / Kimi

ReviewsShips without manual review; stops only for high-risk changes

Quinn Slack

Co-founded Amp and Sourcegraph

AmpwebAmp Dial: mediumorbs

ReviewsAgent reviews first, he scans diffs for red flags

Dax Raad

OpenCode co-founder, SST

OpenCodeAstraFablephone + server

ReviewsAsks a smarter model to review, via subagents

Nick Baumann

Codex at OpenAI, ex-Cline

CodexappGPT-6 AstraGPT-5.3-Codex-Sparkmonothread

Adam Wathan

Tailwind CSS, Refactoring UI

AmpwebCodexappClaude CodewebSonnet 4.5cloud Orbs

Anthony Fu

Vitest, Slidev, UnoCSS, VueUse

Claude CodeCursorideSonnet 5.5Opus 5Storybook

ReviewsWatches the result: Storybook for agent UI, a pulls.review summary on every PR, then the final diff

Andreas Kling

Ladybird browser, SerenityOS

Claude CodeCodexFrontier modelSSH fleet

ReviewsAdversarial review passes by different models

Bryan Cantrill

DTrace, Oxide co-founder and CTO

Claude CodecliChatGPTOpus 4.8no closed loop

ReviewsSelf-review first: nobody else reviews LLM code until the responsible engineer has

Birgitta Böckeler

Thoughtworks, harness engineering

CursorideClaude Code PicliSonnetOpus 4.8

ReviewsRisk triage: probability, impact and detectability set how much she reads; sensors point her to where

Burke Holland

Works on GitHub Copilot

GitHub CopilotappClaude CodewebLuna MaxOpus 5.5project leads

ReviewsRubber duck review by a second model family; he no longer reads the code

Andrew Gallant

ripgrep, Rust regex, Jiff

CodexcliGPT-5.6 Solsandbox

ReviewsFlags his LLM PRs for careful review, then simplifies them by hand; Codex review catches bugs too

Calvin French-Owen

Segment co-founder, ex-Codex

CodexcliClaude CodecliFable 5GPT-5.6 Solworktrees

ReviewsSpot-checks the architecture; Cursor Bugbot and Codex review every PR

Cat Wu

Claude Code product, Anthropic

Claude CodeappClaude TagOpus 5.5Fable 5.1auto mode

ReviewsClaude code review clears the outer layers; code owners review the core

Chip Huyen

AI Engineering author

CodexClaude CodeFableown runner

ReviewsA second agent reviews the plan; UI work comes with screenshot evidence on the PR

Chris McCord

Creator of Phoenix, Phoenix.new

Claude CodeClaudeone convo

ReviewsRules file names the tests to run before claiming a fix

David Crawshaw

exe.dev, Tailscale co-founder

ShelleywebOpenAI modelsfresh VMs

ReviewsReads agent code before it hits serious systems, trimmed by meat

Claire Vo

ChatPRD founder, How I AI host

CodexappClaude CodeCursorwebGPT-6 AstraOpus 5.5

ReviewsCross-model review: Opus 5.5 reviews GPT's PRs and vice versa; PRs also get human review

Dan Luu

danluu.com essays, ex-Centaur

CodexcliGPT-5.6 Solone-off loops

ReviewsHeavy randomized testing instead of human review

Drew Breunig

"How to Fix Your Context", Plumb

Claude CodeFableGLM 5.2

ReviewsSteps through the decisions Plumb extracts at each commit instead of reading the diff

Dex Horthy

HumanLayer CEO, 12-Factor Agents

HumanLayerideCodexClaude CodeGPT-5.6 SolFable

ReviewsStill reads the code, one 1k-3k PR at a time

CodexappClaude CodeGPT-6 SolFableCodex tasks

ReviewsVibe codes; doesn't read the codebase

Alexander Embiricos

Codex product lead, OpenAI

CodexappCodex default modellong-lived thread

ReviewsReads the plan; Codex reviews nearly every PR

Edward Z. Yang

PyTorch at Meta, ghstack

Claude CodecliCodexOpus 5.5Fable 5tmux -CC

ReviewsReads every line, steers with questions and edits

Fabien Sanglard

Game Engine Black Books

Claude CodeideAntigravityidesession per feature

ReviewsReads all the code; LLM criticizes first

Filippo Valsorda

Go cryptography, age, mkcert

Shelley (exe.dev)webClaude CodecliGPT-6 AstraOpus 4.8exe.dev VMs

ReviewsLLM review before any human review

Garry Tan

YC CEO, creator of gstack

Claude CodeappCodexcliFable 5.1Conductor

ReviewsEvery branch gets CEO, Codex, DX and eng review passes

Jeff Geerling

Ansible, Raspberry Pi YouTube

ClaudewebOpus 5.5Fable 5.1chat + files

ReviewsReviews in stages, like a tech lead with juniors

Geoffrey Litt

Malleable software, Notion

Claude CodeCursorideCodex CLIcliOpus 5Fable 5

ReviewsReads an explainer first, then the full diff; must pass a 5-question quiz before sending code on

George Hotz

tinygrad, comma.ai

OpenCodeKimiOpuslocal models

ReviewsReads and understands each line

Jarred Sumner

Creator of Bun, now at Anthropic

Claude CodecliRoboBunFable 5workflows + /loop

ReviewsTwo adversarial Claude reviewers per implementer, then he reads the outputs

Jessica Kerr

Symmathesy, Honeycomb

Claude CodeOpus 5.5Sonnet 5.5subagents

ReviewsA reviewer agent checks each PR against a stated correctness condition; her own reviews fix the system

José Valim

Creator of Elixir, Tidewave

CodexTidewaveappSol 5.6review pane

ReviewsReviews agent PRs like any other code

Jeremy Howard

fast.ai, Answer.AI, ULMFiT

Claude CodecliSolveitwebCodexOpus 4.5persistent kernel

ReviewsSmall steps he follows one by one; the agent must propose before touching his published libraries

Jason Liu

Instructor creator, OpenAI DX

CodexappGPT-5.5GPT-5.3-Codex-Sparkpinned threads

ReviewsReview skills modelled on colleagues' PR feedback; memory changes reviewed as git diffs

Kent C. Dodds

Testing Library, Epic Web

Grok BotCursorwebGrok 4.5Fablecloud agents

ReviewsTrusts gates and AI reviewers, reads the outcome, not the diff

Kenton Varda

Cloudflare Workers, Cap'n Proto

Claude CodecliCloudflare OSwebFable 5.1Solplan first

ReviewsSol reviews every commit; he reviews the plan and the code

Steph Ango

CEO of Obsidian

Claude CodecliObsidian CLI

ReviewsRecorded test output must be read; fixture must fail before the fix

Lee Robinson

ML at SpaceX, ex-Cursor, Vercel

Grok BotappCursorwebOpus 4.5GPT-5.1-Codex-Maxalways-on bots

ReviewsReviews most of the code and tests it; Bugbot reviews every PR

Will Larson

Staff Engineer, CTO at Imprint

Claude CodeAgent FleetCodexGPT 5.4~10 workspaces

ReviewsSame harness reviews first, context cleared; humans after

Maggie Appleton

Digital garden, GitHub Next

Claude CodewebCodexOpenCodeOpus 5Sonnet 4.6

ReviewsWatches the diffs, checks code in an IDE

Alex Kladov

rust-analyzer, TigerBeetle

Claude CodecliClaude 3.7 Sonnetspec + skeleton

ReviewsOwns the structure, runs the code, does not vet the guts

McKay Wrigley

Founder, Takeoff AI

CodexappClaude CodecliOpus 5.5GPT-6 Astraagent swarms

ReviewsApproves without reading the code

Mike McQuaid

Homebrew maintainer since 2009

CodexcliClaude CodecliOpenCodecliFable 5OpenAI Codex 5.4

ReviewsReviews agent work before pushing it himself; depth depends on the code

Mike Krieger

Instagram co-founder, Anthropic Labs

Claude DesktopappOpus 5.5Sonnet 5.5coordinator Claude

ReviewsHands over whole goals and has Claude verify, then double-verify

Max Woolf

Ex-BuzzFeed data scientist

Zed AgentideCodexcliClaude CodecliGPT-6 AstraGPT-5.6 Luna

ReviewsLuna subagents re-check until all satisfied; he watches benchmark diffs

Nathan Lambert

RLHF Book, Interconnects

Claude CodecliCodexFable 5GPT 5 ProDGX Spark

ReviewsLets agents catch their own mistakes, then checks the results himself

Ned Batchelder

coverage.py

Claude CodecliGitHub CopilotideLSP tool

ReviewsA person owns and reviews every commit, Claude or not

Nolan Lawson

PouchDB, emoji-picker-element

Claude CodeCodexcliCursor BugbotwebOpus 4.7GPT 5.5

ReviewsThree agents hunt bugs in every PR; he triages the ranked report and guides the fixes

Orta Therox

Puzzmo, CocoaPods, Danger

Claude CodeideClaude Code on the webwebFableclones

ReviewsReads all Puzzmo code; Claude reviews every PR too

Paul Irish

Chrome DevTools, Lighthouse

CodexwebGitHub Copilot coding agentwebJuleswebGemini 3 Flash (preview)cloud PRs

ReviewsCodex bot reviews every PR; he has Copilot fix the findings

Sebastian Raschka

Build an LLM From Scratch author

CodexClaude CodeQwen CodecliGPT-5.5Qwen3.6-35B-A3B

ReviewsWants to see what the agent is doing; writes the code he cares about himself

Guillermo Rauch

CEO of Vercel

Ryan Dahl

Creator of Node.js and Deno

Claude CodeCodexOpus 4.6standalone VMs

Sean Goedecke

Writing on AI and big tech

GitHub Copilot appappCopilot CLIcliGPT-6-AstraOpus 5.5Copilot app

ReviewsSkims each attempt for ~30 seconds, rejects most, then fully reviews the one he keeps

Sahil Lavingia

Gumroad founder

Claude CodeCodexFable 5GPT-6issue → PR

ReviewsDoes not just hit merge: QA, then asks a colleague to check

Ryan Carson

Founder of Untangle

DevinwebCodexOpus 5.5SWE 1.7Devin + Linear

ReviewsAgents review and land; he watches a test video first

Steve Klabnik

Co-author of the Rust book, jj

Claude CodecliCodexappFableGPT Soljj workspaces

ReviewsCodex reviews Fable's PRs; humans stay responsible for what they submit

swyx

Latent Space, AI Engineer

CodexClaude CodecliDevinwebOpus 4.6Ghostty + cmux

ReviewsDevin Review bot comments on each agent PR, often after merge

Terence Tao

Fields Medal mathematician, UCLA

Claude CodeOpus 5sub-agents

ReviewsChecks results, doesn't read machine code

Thariq Shihipar

Claude Code team, Anthropic

Claude CodecliClaude TagFable 5Opus 5.5Claude Tag

ReviewsClaude writes an HTML report and a quiz on the change; he merges only after passing it

Sebastián Ramírez

FastAPI, Typer, SQLModel

CodexGPT-6 Astra/goal

ReviewsReviews and adjusts all AI code by hand

Thibault Sottiaux

Codex & ChatGPT at OpenAI

CodexappGPT-5.6 SolGPT-6 Astraautomations

ReviewsReviews what the agent presents; uses a fast model to understand code, never delegates understanding

Tom MacWright

Val Town, Placemark, ex-Mapbox

Claude Codecli PiSonnet 5GLM 5.2Neovim

ReviewsCareful human review of LLM fixes

Tobi Lütke

Shopify CEO, qmd

Claude Code PicliHermes AgentOpus 5.5GPT-6 Astra

ReviewsSpec suite the agent must pass

Nikita Prokopov

DataScript, Fira Code

Claude CodeCodexOpus 5.5live nREPL

Linus Torvalds

Creator of Linux and Git

ClaudeOpusside-branch issues

ReviewsOpus writes verification scripts; the AI checks its own output

Thomas Ptacek

Matasano, Latacora, Cryptopals

Claude CodecliCodexOpus 5.5GPT-5.5spare MacBook

ReviewsAgents push PRs, only he merges

Robert C. Martin

Clean Code author

GrokcliCodexcliGrokno harness

ReviewsDoes not read agent code: Codex reviews Grok's output, he inspects a UML diagram colored by CRAP and mutation scores

Walden Yan

Co-founder, Cognition (Devin)

DevinwebFableGPT-5.5Devin teams

ReviewsDevin Review first, then asks Devin what to look at first

Wes Bos

Syntax podcast co-host

Claude CodecliCodexappOpenCodecliOpus 5.5Codex models

ReviewsReviews after the fact: tests it works, checks for duplicated messes

Wes McKinney

pandas, Apache Arrow

Claude CodecliCodexappFableGPT-5.6 Solown tools

Reviewsroborev reviews every commit; he checks structure

Xe Iaso

Anubis, xeiaso.net

Claude CodecliCodexOpenCodeOpus 5.5GPT-6.1 Sol

ReviewsA second agent team reviews; /reviewbot on PRs

Zach Lloyd

Co-founder and CEO of Warp

Warp FactorieswebWarp agentappGPT 5.6 Solfactory-as-code

ReviewsEngineer reviews the agent PR and ships it; QA video attached