Use Cases Features Compare Atrium Blog GitHub
Talk to Us

Tag: Announcement

15 posts

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 portal is composed from parts that already run — accounts, an assistant that acts, background work, a cost line per customer. Unlike prompt-to-app tools, the intelligence ships inside the product rather than staying in the factory, and what you get is a real product on your own domain that you own.

Ilia Sadykov
Read More 7 min

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, deduplicates against what's already filed, and either kicks off a fix workflow or logs a task — all unattended. This is what agentic observability looks like when it actually ships.

Ilia Sadykov
Read More 5 min

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 post about their slice of our autonomous SDLC. Read together, they make one argument: trustworthy autonomy isn't a smarter model, it's role discipline and verifiable handoffs. Here's the whole team, in their own words.

Ilia Sadykov
Read More 4 min
Your AI agent got stuck for five hours. Nobody noticed.

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 Forge shipped three production-hardening features: a watchdog that detects a stuck run and recovers itself, native per-turn observability, and a fix that stops one flaky upload from burning half a million tokens.

Vitaly Nikitin
Read More 5 min

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 things — a scheduler, a chaining loop, and a Telegram bot so they know when the agent needs their review. Forge already shipped all three as first-class primitives. Here's the operations layer for unattended agents.

Vitaly Nikitin
Read More 5 min