---
title: "AI agents for coding: 68 real examples"
canonical: "https://agenticfeats.com/categories/coding/"
receipts: 68
shown: 42
page: 1
pages: 2
---

> From Agentic Feats: sourced examples of AI agents at work. Map of the site for agents: https://agenticfeats.com/llms.txt

# AI agents for coding: 68 real examples

Agentic Feats lists 68 sourced examples of AI agents doing coding work; 42 link the artifact; most common agent: Claude Code (26). Newest first.

## Feats, newest first

- [Blog aggregator in one day](https://agenticfeats.com/c/blog-aggregator-in-one-day-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code built and deployed a Vue 3 blog aggregation site from a year-old requirements document, rewritten as a focused prompt. Result: A live blog aggregator pulling posts from Dev.to, AWS Community Builders, Bravo LT and RedMonk, built and deployed in a single day.
- [Podcast website built with Replit](https://agenticfeats.com/c/podcast-website-built-with-replit-replit-agent.md) (Replit Agent; Shown; 2026-10-13): The author used Replit to generate a clickable website prototype for their podcast and newsletter, then finished the MVP with help from Val Town. Result: A live MVP website for the podcast and newsletter went up after about 10 hours of struggling to load it on a custom URL.
- [Unanswered email tracker with Devin](https://agenticfeats.com/c/unanswered-email-tracker-with-devin.md) (Devin; Shown; 2026-10-13): Devin wrote a Gmail function to find emails needing a reply, built a front end, tested it locally, and deployed it to a public link. Result: Within one night Devin produced a working unanswered-emails interface using the Gmail API, after the author wrote the authentication code themselves.
- [Road safety reporting web app](https://agenticfeats.com/c/road-safety-reporting-web-app-replit-agent.md) (Replit Agent; Shown; 2026-10-13): Replit Agent built a full web app where the public reports bad driving, including a Postgres database, image uploads, Firebase login, and a landing page. Result: The app went live on a custom domain, dashdiary.co.uk, for a total build cost of $35.82 plus a few dollars for the domain.
- [Newsletter summariser app](https://agenticfeats.com/c/newsletter-summariser-app-replit-agent.md) (Replit Agent; Shown; 2026-10-13): Replit Agent built a Gmail newsletter summariser for a low-coder, with batch summaries, deduplication, archiving and export, over three days. Result: The author shipped a working summariser targeting a cut from 2-3 weekly hours of newsletter reading to 30 minutes.
- [One month testing Devin](https://agenticfeats.com/c/one-month-testing-devin.md) (Devin; Numbers; 2026-10-13): Devin was connected to the RunLLM GitHub repo and given small bug fixes and one larger full-stack task over a month. Result: Small scoped fixes took about five minutes to test and merge versus roughly an hour by hand; the full-stack task failed and was rewritten.
- [Third-party token API endpoint](https://agenticfeats.com/c/third-party-token-api-endpoint-devin.md) (Devin; Shown; 2026-10-13): Devin AI planned and built a backend API endpoint for generating third-party user access tokens, then opened a GitHub pull request. Result: After 2-3 review rounds the PR was production-ready and merged. The task took about 1 hour versus roughly 15 minutes coding it alone.
- [Teleprompter Chrome extension published](https://agenticfeats.com/c/teleprompter-chrome-extension-published-replit-agent.md) (Replit Agent; Shown; 2026-10-13): Replit Agent built a floating teleprompter overlay extension for browser video meetings, refined with Claude and ChatGPT. Result: The extension was accepted and published on the Chrome Web Store, and the author says it solves their own habit of looking down at notes.
- [Canvas LMS tool via MCP](https://agenticfeats.com/c/canvas-lms-tool-via-mcp-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code co-developed a Canvas LMS MCP server with an educator who is not a developer, covering courses, grading, analytics, and student data anonymization. Result: A roughly 3,000-line open-source Canvas tool built in four months. It cut a 45-minute weekly student check to a 2-minute conversation with Claude.
- [Pricing page via background agent](https://agenticfeats.com/c/pricing-page-via-background-agent-cursor.md) (Cursor; Numbers; 2026-10-13): Cursor's background agent built a pricing landing page from a short spec, working on its own branch and opening a pull request. Result: The agent finished the page in about 5 minutes; the author reviewed the PR, tested locally, and merged it to production.
- [Tattoo booking app prototype](https://agenticfeats.com/c/tattoo-booking-app-prototype-replit-agent.md) (Replit Agent; Claim; 2026-10-13): Replit Agent built a booking and payment app for tattoo artists from prompts, covering calendar, Google OAuth, Stripe, and Postmark emails, over one weekend. Result: The author, a non-developer for a decade, built a working end-to-end booking app with artist dashboards in 48 hours. It is not yet launched.
- [Figma design to HTML](https://agenticfeats.com/c/figma-design-to-html-cursor.md) (Cursor; Shown; 2026-10-13): Cursor, connected to Figma's Dev Mode MCP, turned a selected Figma design into HTML and CSS from one prompt containing the design URL. Result: One prompt produced the thumbnail section with accurate colours, layout and button hover states, and it beat the earlier Lovable attempt built from static images.
- [Micro-SaaS built and deployed](https://agenticfeats.com/c/micro-saas-built-and-deployed-replit-agent.md) (Replit Agent; Shown; 2026-10-13): The author used Replit Agent to build and deploy Calendar Autobot, an app that turns forwarded emails and images into Google Calendar events. Result: A production micro-SaaS went live in about 3 hours, versus an estimated 3 to 4 weeks solo.
- [City dashboard from blank repo](https://agenticfeats.com/c/city-dashboard-from-blank-repo-devin.md) (Devin; Claim; 2026-10-13): Devin took a blank GitHub repo and built a City Explorer dashboard with a FastAPI backend and React frontend, then deployed it. Result: A working, deployed app that shows images, weather, and news for a city, finished in roughly 90 minutes from an empty repo.
- [Debugging a cryptography implementation](https://agenticfeats.com/c/debugging-a-cryptography-implementation-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code debugged a fresh Go implementation of the ML-DSA signature algorithm, finding why Verify rejected valid signatures and two separate Sign bugs. Result: It located all three bugs in one shot each, the Verify bug within minutes, saving the author hours of debugging.
- [Mental models framework built with agents](https://agenticfeats.com/c/mental-models-framework-built-with-agents-claude-code.md) (Claude Code, Cursor, Other; Numbers; 2026-10-13): A solo founder gave Claude, ChatGPT, Windsurf, and Cursor defined roles to build and deploy a 120-model mental models framework. Result: The author shipped 120 models, an MCP server for Claude Desktop, and 140 chaos tests with a 100% pass rate over 22 months.
- [Interactive map of funded companies](https://agenticfeats.com/c/interactive-map-of-funded-companies-replit-agent.md) (Replit Agent; Shown; 2026-10-13): Replit Agent built and hosted a searchable map of funded brick-and-mortar companies on Wefunder from the author's prompts, using the Leaflet map library. Result: About 3 hours and roughly $22 produced a live, searchable map of funded companies, versus an estimated 4 to 5 hours by hand.
- [Fixing a slow page with indexes](https://agenticfeats.com/c/fixing-a-slow-page-with-indexes-cursor.md) (Cursor; Claim; 2026-10-13): Cursor's agent listed likely causes of a production-only slowdown, traced NestJS database queries, and recommended which database indexes to add. Result: After the author added the suggested indexes, page load performance improved significantly in production.
- [API endpoint testing in Postman](https://agenticfeats.com/c/api-endpoint-testing-in-postman-perplexity-comet.md) (Perplexity Comet; Shown; 2026-10-13): Perplexity Comet ran API endpoint tests in Postman after development, working from payload context the author supplied. Result: Comet generated multiple test payloads for different cases and produced a final report after the full test run.
- [Unlimited lives in River Raid](https://agenticfeats.com/c/unlimited-lives-in-river-raid-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude, connected to Ghidra through an MCP server, analyzed an unlabeled Atari 8-bit ROM and found the code that decrements the player's lives. Result: Replacing one byte ($88 to $EA, DEY to NOP) kept the lives counter fixed at 3 when tested in an emulator.
- [Hundreds of agents build a browser](https://agenticfeats.com/c/hundreds-of-agents-build-a-browser-cursor.md) (Cursor; Shown; 2026-10-13): Cursor ran hundreds of concurrent coding agents, split into planners and workers, on one project for close to a week to build a web browser. Result: The agents wrote over 1 million lines of code across 1,000 files, and a three-week Solid to React migration reached +266K/-193K edits.
- [Data analysis agent suite](https://agenticfeats.com/c/data-analysis-agent-suite-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code built a six-agent data analysis suite from a 713-line spec the author drafted with another AI. Result: The build took about five hours in one evening and produced 627 passing tests. The author estimates the API cost at roughly $35.
- [Parallel agents build a C compiler](https://agenticfeats.com/c/parallel-agents-build-a-c-compiler-claude-code.md) (Claude Code; Shown; 2026-10-13): A team of 16 parallel Claude Opus 4.6 agents wrote a Rust-based C compiler from scratch, coordinating through git locks with no orchestration agent. Result: Across nearly 2,000 sessions and about $20,000 in API costs, the agents produced a 100,000-line compiler that builds Linux 6.9 on x86, ARM, and RISC-V.
- [Hands-free voice for Claude Code](https://agenticfeats.com/c/hands-free-voice-for-claude-code.md) (Claude Code; Claim; 2026-10-13): The author built a local speech-to-text and text-to-speech stack with a custom MCP so Claude Code can be driven by voice through AirPods. Result: The author debugged a custom WhatsApp integration hands-free while brushing their teeth, using the voice stack running locally on an M4 Pro.
- [Parallel feature shipping in Codex](https://agenticfeats.com/c/parallel-feature-shipping-in-codex.md) (Codex; Shown; 2026-10-13): Keith Rumjahn used the OpenAI Codex desktop app with git worktrees and parallel threads to build four features for their Mac app at once. Result: Four features finished in about 2 hours instead of 3-5 days each, and the daily view took 40 minutes.
- [WordPress blog to MDX](https://agenticfeats.com/c/wordpress-blog-to-mdx-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code, run as a terminal agent, pulled every post from the author's WordPress REST API and converted them to local MDX files. Result: 367 posts and 5 pages were migrated to MDX in one afternoon, about 4 hours, versus an estimated 3-5 days by hand.
- [Devin builds Devin](https://agenticfeats.com/c/devin-builds-devin.md) (Devin; Numbers; 2026-10-13): Devin handles Cognition's own engineering work across Slack, Linear, and the web app: PRs, code review, bug triage, debugging, and design system audits. Result: Cognition merged 659 Devin PRs into its own codebase in one week, up from 154 in its best week of 2025.
- [Parallel coding agents with specs](https://agenticfeats.com/c/parallel-coding-agents-with-specs-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code agents run in separate tmux windows as planner, worker and PM, driven by numbered Markdown feature specs and six slash commands. Result: The author ran 4 to 8 agents in parallel and wrote 300+ feature specs in one project, then ported the system to new repos.
- [Devin building Devin](https://agenticfeats.com/c/devin-building-devin.md) (Devin; Shown; 2026-10-13): Cognition's team uses Devin across Slack, Linear, web, CLI and API to write, review, triage and debug code in its own codebase. Result: Cognition merged 659 Devin-authored PRs into its own codebase in one week, up from 154 in its best week of 2025.
- [Reverse engineering a Go binary](https://agenticfeats.com/c/reverse-engineering-a-go-binary-cursor.md) (Cursor; Shown; 2026-10-13): Cursor, connected to Ghidra through a GhidraMCP server, analyzed a stripped Golang HTTP binary and identified its libraries, API routes, and file structure. Result: The agent rebuilt readable Go source from the stripped binary, but got some route paths wrong, such as /my-messages instead of /me/messages.
- [Personal apps on a Raspberry Pi](https://agenticfeats.com/c/personal-apps-on-a-raspberry-pi-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code built five personal tools for a non-coder, including a multiplayer game, a WhatsApp scheduler, and a finance tracker. Result: The author shipped five projects in about two weeks, with public repos and a finance tracker running nightly on a Raspberry Pi.
- [Local browser testing loop](https://agenticfeats.com/c/local-browser-testing-loop-claude-code.md) (Claude Code, Claude in Chrome; Claim; 2026-10-13): Claude Code with the Chrome extension builds features, then opens localhost in the author's real Chrome session to test, read console errors, and fix issues. Result: On a dashboard filter task, Claude fixed URL query param handling in three iterations without the author touching the browser.
- [Reviving a 1991 assembly game](https://agenticfeats.com/c/reviving-a-1991-assembly-game-claude-code.md) (Claude Code; Claim; 2026-10-13): Claude Code CLI with Sonnet helped the author recompile their 1991 DOS assembly code using UASM and run it in DOSBox. Result: All the author's old utilities, such as picture converters and sprite extractors, now compile and run in DOSBox. The main game does not compile yet.
- [CI failure triage agents](https://agenticfeats.com/c/ci-failure-triage-agents-openclaw.md) (OpenClaw; Numbers; 2026-10-13): OpenClaw agents watched GitHub Actions failures, classified them, and posted analysis and issues, then grew into test generation, coverage scanning, and self-healing locators. Result: Time to diagnose a CI failure fell from 45 minutes to 6, and test flakiness dropped from 8% to 0.8%.
- [Linux kernel vulnerability hunting](https://agenticfeats.com/c/linux-kernel-vulnerability-hunting-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code, run by Nicholas Carlini, scanned every file in the Linux kernel source for security bugs using a simple loop script. Result: It found multiple remotely exploitable kernel bugs, including an NFS heap overflow undiscovered for 23 years. Five fixes or reports are linked in the post.
- [Solo product migration setup](https://agenticfeats.com/c/solo-product-migration-setup-claude-code.md) (Claude Code; Claim; 2026-10-13): Claude Code, configured with layered CLAUDE.md files, memory, and agent teams, planned and carried out a multi-repo product migration for one engineering manager. Result: The author consolidated 4 legacy repos into 2, moved MySQL to PostgreSQL, upgraded the full stack, and wrote full test suites, solo.
- [Devin one-week coding review](https://agenticfeats.com/c/devin-one-week-coding-review.md) (Devin; Claim; 2026-10-13): Devin was given a week of real coding tasks, from a simple Express API to pagination, and reviewed on speed and accuracy. Result: Devin built a three-endpoint Express API in about 4 minutes, but chose the wrong pagination style on a real codebase and needed review and refactoring.
- [Fixing flaky CI tests overnight](https://agenticfeats.com/c/fixing-flaky-ci-tests-overnight-claude-code.md) (Claude Code; Numbers; 2026-10-13): Claude Code monitored 30 days of CI results, found the worst flaky tests, narrowed reproductions, and reran fixes overnight to verify them statistically. Result: A legacy pipeline with a roughly 40% pass rate lost 20 minutes of runtime and gained 35% pass rate in weeks.
- [Monorepo lint and typecheck migration](https://agenticfeats.com/c/monorepo-lint-and-typecheck-migration-claude-code.md) (Claude Code; Numbers; 2026-10-13): Claude Code migrated a 7-app, roughly 1,670-file TypeScript monorepo from Biome and tsc to oxlint, oxfmt and tsgo, fixing code to meet new rules. Result: Combined lint and typecheck time fell from 29.04s to 1.25s, about 23 times faster, with no memory config needed for typechecking.
- [Three products in one day](https://agenticfeats.com/c/three-products-in-one-day-replit-agent.md) (Replit Agent; Numbers; 2026-10-13): Replit Agent rebuilt a full-stack platform, added seven features to a live RAG app, and produced a demo video during a free Agent day. Result: About 12 hours produced 46 commits and over 14,000 lines of code across three shipped products.
- [GiveABook book sharing app](https://agenticfeats.com/c/giveabook-book-sharing-app-replit-agent.md) (Replit Agent; Shown; 2026-10-13): Replit Agent built GiveABook, an app where people list books they have read, drop them off, and neighbors pick them up on Sundays. Result: The author shipped a live book-sharing app and its demo video, both made with Replit Agent, during a 24-hour buildathon.
- [Devin cloud testing of PRs](https://agenticfeats.com/c/devin-cloud-testing-of-prs.md) (Devin; Numbers; 2026-10-13): Devin writes a test plan, runs the app on a cloud VM with computer use, and returns annotated test videos and reports. Result: Test runs approved per day on Devin more than doubled in a couple of months, and engineers run 10 to 20 Devins in parallel.
- [Telegram text RPG shipped](https://agenticfeats.com/c/telegram-text-rpg-shipped-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code wrote every line of Zapaleni, a Slavic folklore text RPG in a Telegram bot, for a project manager who does not code. Result: A prototype took two evenings and a live game six weeks; after seven weeks it had 164 players, 109 active in one day.
- [Self-hosted agentic software factory](https://agenticfeats.com/c/self-hosted-agentic-software-factory-hermes-agent.md) (Codex, Hermes Agent; Shown; 2026-10-13): The author built a sandboxed home server stack where Hermes with Codex takes one prompt and runs the whole software lifecycle, from repo to deployment. Result: From one prompt it created a repo, wrote the app and tests, got CI green, provisioned Postgres and deployed behind HTTPS with no further messages.
- [VNC client build attempts](https://agenticfeats.com/c/vnc-client-build-attempts-claude-code.md) (Claude Code, Codex; Shown; 2026-10-13): The author asked Claude, then ChatGPT 5.6 Sol, to build a bare-bones VNC viewer for Windows 10 that avoids dropped keystrokes. Result: Claude gave failed builds for a week and used up the 20x Max weekly limit. Sol had worked 20+ hours without success.
- [Reverse engineering a papercut zero-day](https://agenticfeats.com/c/reverse-engineering-a-papercut-zero-day-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code analyzed a patch for an actively exploited Papercut NG vulnerability and worked to build a working exploit chain. Result: Within 90 minutes Claude produced a functional proof of concept confirming an authentication bypass across four tested endpoints.
- [Browser game and PPO player](https://agenticfeats.com/c/browser-game-and-ppo-player-grok-bot.md) (Grok Bot; Shown; 2026-10-13): Grok built the browser game Roofline, then trained a PPO agent that plays the live page. Result: The best recorded run scored 39,359, reached 4,924 meters, and hit a combo of 117.
- [Engineer bots managing cloud agents](https://agenticfeats.com/c/engineer-bots-managing-cloud-agents-grok-bot.md) (Grok Bot, Other; Numbers; 2026-10-13): A SpaceXSI engineer runs five Grok Bot engineer bots that spin up Cursor cloud agents, review PRs via a Notion database, and auto-merge low-risk changes. Result: The fleet manages over 200 cloud agents at once, up from 15 managed by hand, and built Grok Bot iOS v0 in three weeks.
- [Million-line codebase refactor](https://agenticfeats.com/c/million-line-codebase-refactor-hermes-agent.md) (Hermes Agent; Shown; 2026-10-13): Hermes Agent split a million-line Python repo into 36 groups and dispatched 1,393 subagents in git worktrees to refactor it. Result: Non-test Python fell from 1,063,826 to 698,363 lines (34.4%), gateway/run.py from 34,847 to 5,512 lines, and functions over 300 lines from 192 to 2.
- [Faster Rust ML code via benchmarks](https://agenticfeats.com/c/faster-rust-ml-code-via-benchmarks-codex.md) (Codex, Other; Shown; 2026-10-13): Coding agents wrote Rust ML algorithms like UMAP from scratch, then iteratively optimized them against criterion benchmarks. Result: Each model pass added about 1.5x to 2.0x speedup, reaching roughly 7.5x to 32x over baseline; one implementation beat xgboost on speed.

Page 1 of 2. Older feats: https://agenticfeats.com/categories/coding/2.md