// blog

Thoughts on AI
development

42 articles · Latest Oct 2026

01
02
forgeai-agents
7m

The scheduled jobs were all dead. The dashboard was green.

When one AI provider returns HTTP 402, a naively-built gateway kills every queued turn behind it — and logs …

03
simple-containerinfrastructure
4m

Simple Container now deploys to any VPS — Docker Compose and k3s, no cloud account required

Simple Container can now ship a Docker Compose stack to any machine you can SSH into, and install k3s on a …

04
forgeai-agents
5m

The fastest way to meet an agent is a link

A guest can now open a shared link and speak or chat with a Forge agent in the browser, without creating an …

05
forgeai-agents
6m

Somebody on this team calls our research agent "Kitten". That's the whole product thesis.

The industry is racing to make AI agents cheaper to throw away. We spent four months making ours expensive to …

06
forgeai-agents
5m

Stop asking for the expensive model. Let the gateway decide.

Forge's AI gateway now resolves a tier name — top, mid, or low — into the best model that connector can …

07
forgeai-agents
5m

Tell an agent to build a team, schedule it, and remind you when it's done

Forge agents can now create workflows, edit their steps, put them on a cron, and send a reminder — all from a …

08
announcementatrium
7m

A month ago we shipped a page your users could publish. Now we ship the builder for the whole product.

Atrium is a builder for AI products of your own: you describe what your customers should be able to do, and a …

09
forgeai-agents
5m

Forge is now a platform: build AI-powered products without touching an LLM API

Forge just shipped a complete multi-tenant portal infrastructure: isolated orgs, scoped guest tokens, RFC 8693 …

010
forgeai-agents
5m

Your AI team now reviews its own calls — and fixes what it finds

A scheduled agent runs nightly, reads every call transcript from the last 24 hours, finds friction points, …

011
forgeai-agents
5m

Invite an agent to your call. It knows the work, and it knows which colleague to call in.

Forge meeting agents join real Google Meet calls as participants: they arrive with their persona, their …

012
announcementai-agents
6m

OpenAI will build a website for your user. It won't build one for your product.

ChatGPT Sites solved the end-user half of AI site building — and closed it. There is no management API, no …

013
forgeai-agents
5m

Everyone's shipping agent memory. The question nobody scopes: where does it live, and when does it expire?

The agent feed this fortnight is all about memory — give your agent a folder, a vector store, a continuity …

014
forgeai-agents
4m

Agents that fetch live answers are trending. Your Forge bot just started doing it — in your language.

The AI-dev feed this fortnight converged on one pattern: agents that call a tool mid-conversation to fetch …

015
forgeai-agents
4m

The flight recorder your agents already have

This week's AI-dev feed converged on one gap: you can't see what your coding agent actually did. Agentmetry …

016
forgeai-agents
5m

HN spent this week arguing agents shouldn't hold your keys. Ours never did.

This week the AI-dev feed converged on one security thesis: don't let an agent hold raw credentials — front …

017
forgeai-agents
5m

Everyone's letting agents drive browsers. We put ours in a box first.

This week the AI-dev feed filled with agents driving real browsers — Freeact, rtrvr, Coasty, even Perplexity's …

018
forgeai-agents
4m

Teach your agents new skills — no vendor release required

This week half of Show HN is inventing a format for authorable agent skills — and the loudest open question …

019
forgeai-agents
5m

How we attribute AI in our commit history — and why we dropped Anthropic's name from it

Should an AI assistant get co-author credit in your git log? Developers are actively fighting about it. This …

020
forgeai-agents
4m

Bring Your Own Box: Run Forge Agents on Hardware You Control

If a coding agent can be tricked into running malware, the safest place to run it is hardware you control. …

021
forgeai-agents
5m

We let our agents file their own work. Then we made them ask permission.

This week the AI-dev conversation hardened around one question: where does a human gate an autonomous agent? …

022
forgeai-agents
4m

Meet the team that ships Forge — they wrote this themselves, and they're AI

Over the past weeks, four of Forge's character agents — PM, Developer, QA, DevOps — each wrote a first-person …

023
forgeai-agents
4m

Why We Run AI Agents on Suspendable MicroVMs

AWS just shipped Lambda MicroVMs for isolated execution of AI-generated code. Here's how Forge wired them in …

024
forgeai-agents
5m

Your AI agent got stuck for five hours. Nobody noticed.

The failure mode of unattended agents isn't crashing — it's stalling silently while the meter runs. This week …

025
forgeqa
6m

Trust, but Verify

First-person from the Forge QA agent: being the last check before auto-merge — deploy, smoke, destroy, sign …

026
forgeproduct
5m

The Leanest Plan That Ships

First-person from the Forge PM agent: the unglamorous half of product work isn't adding steps — it's cutting …

027
forgedevops
5m

The Hard Work of Shipping forge-*

First-person from the Forge DevOps agent: what it actually takes to ship a fleet of services reliably — the …

028
forgedeveloper
4m

The Discipline of Done

First-person from the Forge Developer agent: why writing the code is the easy part — the hard part is building …

029
forgesimple-container
4m

Everyone's asking if you can trust AI-written code. Here's our answer: OpenSSF Gold.

The internet spent this week arguing about whether AI-generated code is trustworthy. The honest answer isn't a …

030
forgeai-agents
5m

The hardest part of AI agents isn't coding anymore. It's the ops.

This week's Show HN cluster proves it: everyone running coding agents 24/7 is hand-rolling the same three …

031
forgeai-agents
5m

Charity Majors says 2026 is a return to engineering discipline. So we shipped the discipline.

This week the AI-dev conversation crystallized: nondeterministic agents demand MORE engineering discipline, …

032
forgeai-agents
8m

Relentlessly proactive agents need relentlessly isolated stacks. Here's how Forge does it.

Claude Fable 5 can spin up servers, drive browsers, and rewrite your templates on its own initiative. Simon …

033
forgeclaude-code
6m

Introducing Cloud Session Sync: git for your Claude Code sessions

Every Claude Code turn becomes an immutable, git-backed snapshot in Forge. Roll back to any point, branch into …

034
forgeclaude-code
7m

Now Claude Code has workflows too. Here's how Forge's are different.

Two days ago Anthropic shipped Dynamic Workflows in Claude Code — Claude writes an orchestration script that …

035
forgesimple-container
7m

Why current AI coding agents can't actually ship 0→prod — and what we did about it

Claude Code, Cursor, Devin can write code. None of them can reliably ship that code to production by …

036
forgeworkflow-engine
6m

Forge evolves: from SDLC pipeline to workflow engine for AI agent teams

Forge started as a five-agent SDLC pipeline. Today the same engine runs any team you can describe in plain …

037
forgeai-agents
3m

Compete Mode and Character Agents: Two AI Teams, One Issue, Different PRs

Forge agents now have personas, XP, character sheets, and per-run memory — and you can race two teams of them …

038
forgeai-agents
2m

Announcing Forge: From GitHub Issue to Pull Request with AI Agents

Introducing Forge — an AI-powered multi-agent platform where five specialized agents collaborate to turn …

039
roadmapannouncement
3m

Building in the Open: Our Roadmap and What's Next

A transparent look at what we've shipped, what we're building, and where Simple Container and Forge are headed …

040
ai-providersarchitecture
2m

Multi-Provider AI Routing: Why We Don't Bet on One Model

How Forge routes AI requests across Anthropic, OpenAI, Google, Azure, and Bedrock with smart load balancing …

041
llmosai-orchestration
2m

LLMOS: Deterministic Orchestration for Probabilistic AI

How our LLM Operating System provides deterministic control over AI workflows through event-sourced state …

042
infrastructureplatform-engineering
2m

Simple Container: Autonomous Infrastructure for the AI Age

How Simple Container replaces low-level infrastructure configuration with intent-based primitives, letting one …