Coding agents have moved well beyond autocomplete. Tools that can read a whole repository, plan a change, edit many files and run the tests are now part of everyday development for a lot of people. The question is no longer whether they work, but which tasks they handle well.
This guide collects real, sourced examples from developers using coding agents on actual projects: framework migrations, bug hunts, refactors, test writing, prototypes and side projects built mostly by an agent. Each case names the tool used and links to the original write-up.
Reading through them, a few habits come up again and again. Developers who get good results ask for a plan before any edits, point the agent at official documentation, keep changes on a separate branch, and review the diff and test output before merging. The agent does the typing and searching; the developer stays responsible for the design.
Use the cases below as a menu of tasks worth delegating. Open one that matches your work, copy its reusable prompt and adapt it to your codebase. Starting with a well-tested, low-risk area is the quickest way to learn where an agent fits in your workflow. The list is ordered by interest.
What is in this guide: 60 sourced cases, added between 2026-10-30 and 2026-11-14. The tools that come up most often are Claude Code (8), Amazon Bedrock (8), ChatGPT (7), and the cases mostly sit under Software Development (58), Developer Tools (2). They were reported on Case study (Anthropic) (13), Case study (OpenAI) (8), Case study (Microsoft) (6). If you only have a few minutes, start with the top three: “A contractor's personal Claude Code agent allegedly ran inside a secure Pentagon system for two months”; “Claude Code's 3-word status line reportedly burns 15 to 20% of tokens”; “Rocket Money on building agents that fix their own code”.
Each case page gives a short summary in our own words, a note on why it matters, a reusable prompt broken into steps you can follow, and a link to the original first-hand account. Nothing here is a paid placement, and we only list cases with a working public source.
Rocket Money on building agents that fix their own code· Claude · Software Development Rocket Money Qa — Read the Q&A to see how they build on the Claude Platform. Q&A | Claude Rocket Money on building agents that fix their own code…
Legora helps lawyers work more efficiently with Claude· Claude Code · Software Development Legora uses Claude throughout their platform to power assistant tools, document review, and intelligent workflows. Legora scored 18% higher on its…
LG CNS modernizes 20-year-old enterprise systems with Claude· Amazon Bedrock · Software Development Claude Code is the engine of its newest offering: LG CNS Build Factory, a turn-key software development service that modernizes decades-old…
Brex reimagines spend management with Claude in Amazon Bedrock· Amazon Bedrock · Software Development Brex automates 75% of expense transactions at 94% compliance. Now, Brex automates 60% of expenses, saving customers about 73,600 hours monthly with…
Office Hours: Building for the model that doesn’t exist yet· Claude Code · Software Development Ramp Boris Office Hours — Boris Cherny sat down with CTO Rahul Sengottuvelu and Staff Software Engineer Austin Ray to talk about treating agents…
Inscribe cut time spent on fraud review 20x with Claude· Amazon Bedrock · Software Development Inscribe cut fraud review from 30 minutes to 90 seconds. Conor Burke, CTO and co-founder of Inscribe, articulates their mission: “We aim to create…
Genspark’s Super Agent orchestrates 150+ tools with Claude· Claude Code · Software Development Genspark — The product is built around the Super Agent, which uses Claude to coordinate more than 150 specialized tools in response to whatever a…
How Vercel Uses Notion to Scale Launches Across the Company· Notion AI · Software Development Vercel Uses Notion to Scale — Its unified system has resulted in 89% of employees reporting increased confidence in the quality of their shipped…
Warp’s big bet on building open source with GPT-5.5· OpenAI · Software Development GPT-5.5 — When Warp [open-sourced(opens in a new window)](https://www.warp.dev/blog/warp-is-now-open-source) its terminal client this year, with…
Abnormal AI scales autonomous development with CodeRabbit· CodeRabbit · Software Development Abnormal AI uses CodeRabbit to scale autonomous development, according to the official case study describing how the company put the AI code-review…
Why AI could be the best medicine for the NHS· Microsoft 365 Copilot · Software Development A ground-breaking pilot of Microsoft 365 Copilot across 90 NHS organisations has found that AI-powered administrative support could save NHS staff…
How Ramp engineers accelerate code review with Codex· ChatGPT · Software Development At Ramp, engineers are using Codex with GPT‑5.5 to accelerate code review and develop internal agentic tooling, helping teams get substantive pull…
How GPT-5.6 Sol helps run quantum computing experiments· ChatGPT · Software Development The results suggest that current agents can handle clearly defined experimental workflows, but interpreting ambiguous physical results remains a…