# RobotPaper

> Roy Osherove's blog on AI patterns and anti-patterns, AI and coding agents, and engineering leadership.

RobotPaper (https://robotpaper.ai) is the personal blog of Roy Osherove: long-form, hands-on posts about AI and coding agents, patterns and anti-patterns for building with LLMs, agent infrastructure on AWS, developer experience, and engineering leadership. Posts often include code and configuration.

Author: Roy Osherove - Senior Solutions Architect at AWS working on AI for production and AI-driven development; author of The Art of Unit Testing and Elastic Leadership, with about 20 years of consulting and training on unit testing, TDD, team leadership and continuous delivery.

Licensed under CC BY 4.0 — reuse, quote and adapt with attribution (Roy Osherove, link to the original post). License: https://creativecommons.org/licenses/by/4.0/

Every post and page is also available as Markdown at its URL with `.md` appended (the HTML page stays canonical). This file is regenerated automatically when posts change; last generated 2026-10-23T08:15:02.942Z.

## Author

- [About Roy Osherove](https://robotpaper.ai/about-roy.md): bio, sites, socials and contact as Markdown
- [RobotPaper](https://robotpaper.ai/): This blog: AI patterns and anti-patterns, AI and coding agents, and engineering leadership.
- [osherove.com](https://osherove.com): Roy's personal site. Author of The Art of Unit Testing and Elastic Leadership; consulting and training on unit testing, TDD, team leadership and continuous delivery, plus about 20 years of blog posts.
- [The Art of Unit Testing](https://www.artofunittesting.com): The Art of Unit Testing, 3rd Edition (Manning): readable, maintainable, trustworthy unit tests, plus TDD workshops in C#, JavaScript, Java and Kotlin.
- [5whys.com](https://www.5whys.com): Roy's writing on software team leadership: self-organizing teams, surviving chaos, learning mode and the Elastic Leadership model.
- [Pipeline Driven](https://pipelinedriven.org): Continuous delivery and DevOps practices: build patterns, pipeline-friendly testing, DevOps metrics and enterprise DevOps.
- [Elastic Leadership](https://www.elasticleadership.com): Elastic Leadership, Roy's framework for software team leaders (survival, learning and self-organizing modes).
- [X](https://x.com/royosherove)
- [LinkedIn](https://www.linkedin.com/in/osherove/)
- [Contact form](https://forms.gle/jQSx5KoXvrVA1JCK7)

## Usage

- [License: CC BY 4.0](https://creativecommons.org/licenses/by/4.0/): reuse, quote and adapt with attribution (Roy Osherove, link to the original post)
- [Markdown version of any post](https://robotpaper.ai/the-new-bottlenecks-after-coding.md): append .md to any post or page URL; the HTML page stays canonical
- [Home page as Markdown](https://robotpaper.ai/index.md): every post with its date, description and .md link
- [Full text of all posts](https://robotpaper.ai/llms-full.txt): every published post as Markdown in one file

## Developer-Experience

- [Markdown is the new Installer](https://robotpaper.ai/markdown-is-the-new-installer.md): No more MSI or drag-to-install. A well-crafted bootstrap prompt lets your users' local agent install complex tools for them. Why markdown is the new installer.
- [Software Engineering Practices (are also) Useful for Token Reduction](https://robotpaper.ai/software-engineering-practices-are-also-useful-for-token-reduction.md): Good engineering practices used to save human cycles. Now person cycles are agent cycles, measured in tokens. Why clean code, tests and DRY cut your token bill.
- [Useful Bash Scripts](https://robotpaper.ai/useful-bash-scripts.md): A great collection of small, useful bash scripts I found, like boop for audible test results and nato for spelling strings out loud. Here are my favorites.
- [AIDHD](https://robotpaper.ai/aidhd.md): AIDHD: when agentic coding makes every side project and coding indulgence just minutes away, focus becomes the real problem. On AI-fueled distraction for devs.
- [GitHub Spec-Kit Spec Driven Dev is in the Right Direction: Disciplined, Incremental Vibe-Coding](https://robotpaper.ai/i-tried-out-github-spec-kit-and-all-i-got-was-this-not-terrible-website.md): I tried GitHub Spec-Kit with Cursor and grok-code-fast to build an app end-to-end: specify, plan, tasks, TDD and Playwright tests. What worked and what didn't.
- [Wanted](https://robotpaper.ai/wanted.md): Wanted: the AI-augmented developer. AI-native engineer? Context engineer? Code orchestrator? A tongue-in-cheek list of every name for the new kind of developer.

## Pattern

- [The New Bottlenecks After Coding](https://robotpaper.ai/the-new-bottlenecks-after-coding.md): Coding got cheap. Now deployment, code review, merge conflicts, architecture context, specs and monitoring are the bottlenecks. Here's my list and some fixes.
- [Pattern: Agent-Owned Accounts (Deploy Environments)](https://robotpaper.ai/agent-owned-accounts-deploy-environments.md): Give an agent a VM and it submits PRs. Give it its own cloud account and it codes, deploys, debugs and monitors apps. The Agent-Owned Accounts pattern.
- [Github Repo Forks Are a New Token-Saving Mechanism](https://robotpaper.ai/github-repo-forks-are-a-new-token-saving-mechanism.md): When you fork an open-source agent repo, you build on someone else's spent token budget. Why written code is now spent tokens, and forks are a way to save them.
- [The Rise of the Stateful Agent](https://robotpaper.ai/the-rise-of-the-stateful-agent.md): OpenClaw kicked off a new pattern: the stateful agent. Persistence, long-term state, proactivity and tools in one loop. Why it matters and where it's going.
- [Stateful Agent — Pattern Definition (with Conversation Log)](https://robotpaper.ai/pattern-stateful-agent.md): A definition of the stateful agent pattern, built in a long debate with Claude Opus 4.6 using Karpathy's argue-the-opposite technique. Full conversation log.
- [Red Queen Code Review Pattern with AI](https://robotpaper.ai/red-queen-code-review-pattern-with-ai.md): Red Queen code review, a pattern by my colleague Harel Coman: every AI code review issue must change either your code or the review rules file.

## Tools

- [Graphlin - Jev-Enabled Live Architecture Visualizer](https://robotpaper.ai/graphlin-jev-enabled-live-architecture-visualizer.md): Graphlin is my open-source Claude Code/Codex plugin that draws a live diagram of the architecture your agent is building, using Jev to classify each action.
- [opstalk - a cli to talk to AWS DevOps Agent](https://robotpaper.ai/opstalk-a-cli-to-talk-to-aws-devops-agent.md): opstalk is a small open-source terminal CLI I built to chat with AWS DevOps Agent: think Claude Code for AWS ops, with scriptable one-shot commands.
- [embedrock : OpenAI-compatible Bedrock embedding proxy for OpenClaw and Friends](https://robotpaper.ai/embedrock-openai-compatible-bedrock-embedding-proxy-for-openclaw-and-friends.md): embedrock is a tiny Go proxy that maps OpenAI's /v1/embeddings API to Amazon Bedrock Titan and Cohere models, for OpenClaw semantic search. Now in bedrockify.
- [RepoSwarm: Give AI Agents Context Across All Your Repos](https://robotpaper.ai/reposwarm-give-ai-agents-context-across-all-your-repos.md): RepoSwarm is an open-source tool that uses Temporal workflows and Claude to analyze hundreds of repos and write architecture docs AI agents can use as context.

## AWS

- [Prodo-typing](https://robotpaper.ai/prodo-typing.md): Prodo-typing: building a production-shaped prototype in its own cloud account with Loki, my OpenClaw-based builder agent. Brainstorming what that changes.
- [Bedrockify: OpenAI Compatible Proxy for Bedrock: Completions + Embeddings](https://robotpaper.ai/bedrockify-openai-compatible-proxy-for-bedrock-completions-embeddings.md): Bedrockify is a fast Go proxy that gives Amazon Bedrock an OpenAI-compatible API for chat completions and embeddings. One line to run; works with OpenClaw.
- [I Gave My Agent its Own AWS Account](https://robotpaper.ai/i-gave-my-agent-its-own-aws-account-and-now-it-codes-deploys-and-debugs-full-stack-apps.md): I gave my agent its own AWS account through Loki, a harness I built. Now it codes, deploys and debugs full-stack apps, with the account as the blast radius.

## Conferences

- [Video: Zero-Touch Algorithm Handoffs: Shipping ML Algorithms to Prod Without Rewrites (or ML Engineers)](https://robotpaper.ai/video-h-zero-touch-algorithm-handoffs-shipping-ml-algorithms-to-prod-without-rewrites-or-ml-engineers.md): My BuildStuff talk on zero-touch algorithm handoffs: shipping ML algorithms to production with Temporal workflows, without rewrites or ML engineer handoffs.
- [Video: RepoSwarm - Giving AI Agents Architecture Context Across All Your Repos (BuildStuff Conference)](https://robotpaper.ai/video-reposwarm-giving-ai-agents-architecture-context-across-all-your-repos-buildstuff-conference.md): My BuildStuff conference talk on RepoSwarm: giving AI agents architecture context across all your repos, with a deep dive into how the tool works and its uses.
- [Upcoming Speaking Engagements (2025)](https://robotpaper.ai/upcoming-speaking-engagements-2025.md): Where I'm speaking in 2025: GenML in Tel Aviv on zero-touch ML algorithm handoffs, and Build Stuff in Vilnius on RepoSwarm and AI agent architecture context.

## Cursor

- [Cursor & Other AI dev tools need a model-switch hook](https://robotpaper.ai/cursor-other-ai-dev-tools-need-a-model-switch-hook.md): We keep forgetting to switch between smart and cheap models mid-chat in Cursor and Claude Code. Here is why AI dev tools need a model-switch hook.
- [A Frontend Tech Lead's Process for AI Assisted Coding in Cursor](https://robotpaper.ai/a-frontend-tech-leads-process-for-ai-assisted-coding-in-cursor.md): My colleague Harel Coman wrote up his AI-assisted coding process in Cursor as a frontend tech lead. It produced great code, and we're rolling it out wider.
- [Onboard a New Developer to a Repository Using Cursor Commands](https://robotpaper.ai/using-cursor-commands-to-onboard-a-new-developer-to-a-repository.md): Use Cursor custom commands like /get-started to onboard a new developer: read the README, install mise, set up the environment and show what to do next.

## ClaudeCode

- [Generating Weird & Beautiful Architecture Diagrams with Claude](https://robotpaper.ai/generating-weird-beautiful-architecture-diagrams-with-claude.md): I had Claude Opus 4.5 generate draw.io XML architecture diagrams from a prompt, then asked for 35+ styles, from AWS re:Invent and blueprint to LEGO and ukiyo-e.
- [Getting Claude Code Up and Running using AWS Bedrock Claude Opus 4.5 Model (Long Term API Keys)](https://robotpaper.ai/getting-claude-code-up-and-running-using-aws-bedrock-claude-opus-4-5-model-no-aws-cli-no-iam-roles-or-users.md): How I got Claude Code running on AWS Bedrock with Claude Opus 4.5 using only environment variables and a long-term API key. No AWS CLI, IAM roles or users.

## MLOps

- [How I (Currently) Map Our AI Tool Stack for Dev, Research, Production and Non-Tech](https://robotpaper.ai/how-i-map-our-ai-tool-stack-for-dev-research-production-and-non-tech-in-2025.md): A bird's-eye map of the AI tools we use: chat, coding tools and agents, research studios, gateways, guardrails, prompt monitoring, agents, model serving.
- [Zero-Touch Algorithm Handoffs: How We Ship ML Algorithms to Prod Without Rewrites (or ML Engineers)](https://robotpaper.ai/zero-touch-algorithm-handoffs-how-we-ship-ml-algorithms-to-prod-without-rewrites-or-ml-engineers.md): How we ship ML algorithms to production without rewrites or ML engineers: researchers own Temporal activities, devs own workflows, and updates go live.

## Models

- [Exploring Karpathy's New NanoChat Project](https://robotpaper.ai/exploring-karpathys-new-nanochat-project.md): Andrej Karpathy released nanochat, a minimal from-scratch full-stack ChatGPT clone training pipeline. I'm training it on Lambda and digging into how it works.

## openclaw

- [Reference Architecture: OpenClaw  (Early Feb 2026 Edition, Opus 4.6)](https://robotpaper.ai/reference-architecture-openclaw-early-feb-2026-edition-opus-4-6.md): I had OpenClaw with Claude Opus 4.6 deep-dive into its own source code and write a detailed architecture doc: memory, sessions, channels, tools and more.

## Optional

- [About this site](https://robotpaper.ai/about.md): This is a blog by Roy Osherove. That's me. I work as a Senior Solutions Architect at AWS, where I get to spend lots of time with customers , helping to design solutions, working on AI stuff, both for production use...
- [RSS feed](https://robotpaper.ai/rss/): latest posts
- [Sitemap](https://robotpaper.ai/sitemap.xml): all HTML URLs