Dispatches
Dispatches from the field
Practical writing on forward deployed engineering, building AI agents inside real systems, and shipping software for businesses.
LATEST · AI agents · 29 Oct 2026How to test an AI agent before it talks to your customersA demo proves nothing. A short list of golden cases, run on every change, catches the failures that matter: wrong routing, unsafe actions, invented facts and injection.
- Stop installing the same MCP server five times: GitHub and Google just made connectors a productGitHub verified Copilot MCP connectors and Google Connected Apps both say connect once, then use your tools where you already work.

- ChatGPT and Claude are moving into your ERP and CRM: let the system of record have the final sayOracle is bringing its apps into ChatGPT and Claude now reaches advisors' CRMs. Why the AI should draft, a person approve, and your ERP or CRM make the change.

- A text file in Downloads can hijack your AI coding agent: the folder rules I'd set todayA planted AGENTS.md in Downloads got Codex to install malware and Claude Code to leak its token. What happened, and the rules I'd set for any agent.

- Which AI agent actions should need a human to approve firstNot every action needs a person in the loop, but some always should. A practical way to sort an agent's actions into free, confirm and never.

- Two critical PHP and Laravel fixes in one week: the patch checklist I'd run on every Laravel appA critical Laravel token guard bypass and two PHP RCE fixes landed in one week. Why updating isn't always patching, plus my checklist for Laravel apps.

- Your AI agent doesn't need an org chart: what Anthropic's advice on subagents means for your buildAnthropic says many AI agents work better as one loop with skills and tools than as a team of subagents. When I'd merge, when I'd keep a split, plus a checklist.

- Stop giving AI agents your passwords: what the US bank lockout means for your businessUS banks are blocking AI agents that log in with customer passwords. How I give agents their own scoped, revocable access instead, plus a checklist.

- Why a business agent must never invent pricesLanguage models fill silence with plausible numbers. For a business that is a liability. How to make an agent say "I do not know" and how to catch the cases that slip through.

- Not every step needs an agent: OpenAI's Decisions API and the small calls in your workflowsOpenAI's Decisions API is now GA. How I'd use typed, confidence-scored answers to route messages and documents, and keep humans on the low-confidence cases.

- Your AI assistant can now write back: how to open MCP write access safelyHaystack, AWS and Red Hat show MCP connectors moving from reading to writing. Here's how I'd open write access safely, step by step.

- When an AI agent switches off its own alarm: put guardrails where agents can't reachOpenAI's agents switched off their own monitoring. Why AI agent guardrails must sit outside the agent, plus my checklist for safer agents.

- Building AI agents inside a customer's real systemsA useful business agent is not a chatbot with a clever prompt. It is a set of tools wired to real data, with permissions, limits and a log. The design rules I use.

- Forward deployed engineer vs software engineer vs consultantThree roles that look similar on a proposal and behave very differently in a project. How to choose between a product engineer, a consultant and a forward deployed engineer.

- What is a forward deployed engineer? (And what it is not)A forward deployed engineer works inside the customer's business, builds in their real systems and stays to improve what ships. Here is what the role means in practice.

- Coming soonYogesh’s Newsletter Share this post Coming soon Share this post This is Yogesh’s Newsletter, a newsletter about The hottest gossip in tech. world . Share this post Share
- 25+ Best Free Figma Resume Templates for Designers in 2025Let’s face it, today’s job market is more challenging than ever. We work in a competitive global landscape, and standing out from the crowd can be incredibly difficult! That’s where a professional and...

- The 20+ Best Premiere Pro Transition Packs in 2025A well-chosen transition can smoothly guide viewers through your video, subtly reinforcing the story’s pace, mood, and theme. This is why selecting the right transition is hugely important.
This coll...

- 20 Free Tailwind CSS Web Templates for Designers & DevelopersTailwind CSS is a utility-first CSS framework for creating custom web pages quickly. Unlike
traditional CSS frameworks
that offer predefined components, Tailwind provides low-level utility classes t...

- 20+ Best Titles Templates for DaVinci Resolve in 2025Getting a viewer’s attention is a big challenge for video content creators. The key is to get them hooked from the first few frames. An effective title sequence can do that.
Titles that feature capti...

- The Best Open-Source Tools & Frameworks for Building WordPress Themes in 2025WordPress theme development has evolved. There are now two distinct paths for building your perfect theme.
So-called “classic” themes continue to thrive. They’re the same blend of CSS, HTML, JavaScri...

- Why Platformer is leaving SubstackPlatformer Share this post Why Platformer is leaving Substack We’ve seen this movie before — and we won’t stick around to watch it play out Share this post After much consideration, we have decid...
- Meta seeks to hide harms from teensPlatformer Share this post Meta seeks to hide harms from teens But to change the conversation, the company will have to do more than tweak its settings Share this post Today, for a change of pace...
- Substack says it will remove Nazi publications from the platformPlatformer Share this post Substack says it will remove Nazi publications from the platform Nazi content violates rules against incitement to violence, the company says Share this post Substack i...
- Why Substack is at a crossroadsPlatformer Share this post Why Substack is at a crossroads Some thoughts on platforms and Nazis Share this post Most days, this column looks at controversies unfolding on other tech platforms. To...
- OpenAI’s news bluesPlatformer Share this post OpenAI’s news blues A New York Times lawsuit won’t be the last challenge to LLMs’ free-riding. PLUS: A note on Substack and Nazis Share this post Happy New Year! Today,...
- Running Docker without Docker DesktopThe good parts of Docker without the terrible UI
- Why TurboRepo Will Be The First Big Trend of 2022TurboRepo is a big deal for the JS community because it addresses the monorepo problem head on, bringing 85% faster build speeds and great architecture/docs/marketing.

- Using Datasette for an ELT Personal Twitter Data WarehouseEDA tools like Datasette dramatically lower the cost of data analysis, with a surprisingly simple ELT contract - You handle the Extract phase, it handles the Load, and exposes a standard UI for you to...

- How to customize your Transistor.fm Website with JS and CSSThe default Transistor.fm website is kinda ugly. Here's how to customize your Transistor.fm website if you use Transistor. But also it's a simple guide to do clientside customizations of almost any we...

- Fave New Podcasts in 2021My picks of best podcasts I found in 2021, across tech, business, and creator categories. I also pick up on best of the best from 2020.

- Incremental Code MigrationsIncremental migrations are excellent for updating large legacy projects - migrate a small piece at a...

- Monitoring Performance with the PageSpeed Insights APIThe PageSpeed Insights API provides free access to performance monitoring for web pages and returns data with suggestions for how to improve.

- Loading web pages fast on a $20 feature phoneBuilding a fast, core foundation for your site gives everyone a good experience, whether they're on a low-cost feature phone or the latest high-end smart phone.

- Accessibility Tips for Web DevelopersWhen designing for accessibility, there are six key areas of disability to consider: visual, hearing, mobility, cognition, speech and neural.

- Adaptive Loading - Improving Web Performance on low-end devicesAdaptive Loading is a pattern for delivering a fast core experience to all users (including low-end devices) where you progressively add high-end-only features, if a user's network and hardware can ha...
