<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>UseAgent engineering blog</title>
    <link>https://useagent.org/blog</link>
    <atom:link href="https://useagent.org/blog/rss.xml" rel="self" type="application/rss+xml" />
    <description>How we run AI coding agents in production: decisions, incidents and numbers from building UseAgent.</description>
    <language>en-us</language>
    <lastBuildDate>Wed, 14 Oct 2026 00:00:00 GMT</lastBuildDate>
    <item>
      <title>AI agent sandbox architecture: keep the control plane outside</title>
      <link>https://useagent.org/blog/agent-sandbox-control-plane</link>
      <guid isPermaLink="true">https://useagent.org/blog/agent-sandbox-control-plane</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Architecture</category>
      <category>Sandboxes</category>
      <description>Our AI agent sandbox architecture: auth, secrets, the event log and the gateway stay outside the sandbox, and engines run inside as replaceable adapters.</description>
    </item>
    <item>
      <title>An AI coding agent in Slack that answers in the thread</title>
      <link>https://useagent.org/blog/ai-coding-agent-in-slack</link>
      <guid isPermaLink="true">https://useagent.org/blog/ai-coding-agent-in-slack</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Slack</category>
      <category>Integrations</category>
      <description>How our AI coding agent in Slack works: mention it in a thread, it works on its own computer and replies there. One ingress door, user mapping, durable replies.</description>
    </item>
    <item>
      <title>Claude Code login expired with many sessions open: the fix</title>
      <link>https://useagent.org/blog/claude-code-login-expired-many-processes</link>
      <guid isPermaLink="true">https://useagent.org/blog/claude-code-login-expired-many-processes</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Claude Code</category>
      <category>Debugging</category>
      <description>Claude Code login expired or &apos;OAuth access token has been revoked&apos; across many sessions? Processes share one token and revoke each other. The cause and the fix.</description>
    </item>
    <item>
      <title>How to run Codex in the cloud with your ChatGPT plan</title>
      <link>https://useagent.org/blog/codex-chatgpt-plan-in-the-cloud</link>
      <guid isPermaLink="true">https://useagent.org/blog/codex-chatgpt-plan-in-the-cloud</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Codex</category>
      <category>Security</category>
      <description>How we run Codex in the cloud with your ChatGPT plan without the login entering the sandbox, one login per shared thread, and warm turns from 12.6 s to 5 s.</description>
    </item>
    <item>
      <title>Building a computer use agent sandbox with a real desktop</title>
      <link>https://useagent.org/blog/computer-use-agent-desktop-sandbox</link>
      <guid isPermaLink="true">https://useagent.org/blog/computer-use-agent-desktop-sandbox</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Computer Use</category>
      <category>Sandboxes</category>
      <description>What it took to make our computer use agent sandbox usable: clicks that land, typing through the clipboard, a desktop started at boot and a real Xorg session.</description>
    </item>
    <item>
      <title>Deploy long-running AI agents without downtime: what broke</title>
      <link>https://useagent.org/blog/deploying-long-running-ai-agents</link>
      <guid isPermaLink="true">https://useagent.org/blog/deploying-long-running-ai-agents</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Deployment</category>
      <category>Reliability</category>
      <description>How we deploy long-running AI agents without downtime: admission drain, runs that survive a restart, a 3.5-minute outage, Caddy POST retries, deploy batching.</description>
    </item>
    <item>
      <title>E2B vs Daytona: running coding agents on both sandboxes</title>
      <link>https://useagent.org/blog/e2b-vs-daytona-agent-sandboxes</link>
      <guid isPermaLink="true">https://useagent.org/blog/e2b-vs-daytona-agent-sandboxes</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Sandboxes</category>
      <category>Infrastructure</category>
      <description>E2B vs Daytona from running Claude Code, Codex and OpenCode on both: measured cold starts, image limits, desktop at boot, idle rules, memory caps, cleanup.</description>
    </item>
    <item>
      <title>Event sourcing for AI agent runs: Postgres is the truth</title>
      <link>https://useagent.org/blog/event-sourced-agent-runs</link>
      <guid isPermaLink="true">https://useagent.org/blog/event-sourced-agent-runs</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Architecture</category>
      <category>Postgres</category>
      <description>Event sourcing for AI agent runs: an append-only Postgres log, one translation for every engine, a one-transaction finalize, and a UI that only reads.</description>
    </item>
    <item>
      <title>How we built an LLM API gateway with per-user spend limits</title>
      <link>https://useagent.org/blog/llm-gateway-per-user-credit</link>
      <guid isPermaLink="true">https://useagent.org/blog/llm-gateway-per-user-credit</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Claude Code</category>
      <category>Security</category>
      <description>Our LLM API gateway with per-user spend limits gives each member $20 of Claude Opus 5.5 without a key: reserve, settle, per-key budgets and a closed allowlist.</description>
    </item>
    <item>
      <title>How we load test AI coding agents: 50 runs at once</title>
      <link>https://useagent.org/blog/load-testing-50-coding-agents</link>
      <guid isPermaLink="true">https://useagent.org/blog/load-testing-50-coding-agents</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Testing</category>
      <category>Reliability</category>
      <description>How we load test AI coding agents: 50 concurrent OpenCode runs on E2B, the fleet caps that had to move, what failed, and the soak storms behind them.</description>
    </item>
    <item>
      <title>Streaming AI agent responses with SSE: one stream per thread</title>
      <link>https://useagent.org/blog/one-sse-stream-per-thread</link>
      <guid isPermaLink="true">https://useagent.org/blog/one-sse-stream-per-thread</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Realtime</category>
      <category>Architecture</category>
      <description>Streaming AI agent responses with SSE broke for us on every reply. The fix: one thread-scoped stream instead of one per run, with Postgres as the truth.</description>
    </item>
    <item>
      <title>An open source Devin alternative, and how it&apos;s built</title>
      <link>https://useagent.org/blog/open-source-devin-alternative</link>
      <guid isPermaLink="true">https://useagent.org/blog/open-source-devin-alternative</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Architecture</category>
      <category>Devin</category>
      <description>UseAgent is an open source Devin alternative: Claude Code, Codex and OpenCode on their own cloud computers, self-hostable. The design, and where Devin leads.</description>
    </item>
    <item>
      <title>OpenCode model not found: why new models fail, and the fix</title>
      <link>https://useagent.org/blog/opencode-model-not-found</link>
      <guid isPermaLink="true">https://useagent.org/blog/opencode-model-not-found</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>OpenCode</category>
      <category>Debugging</category>
      <description>OpenCode model not found on a model OpenRouter lists? OpenCode 1.x froze its bundled models.dev catalog. Why it happens, how we fixed it, and how you can.</description>
    </item>
    <item>
      <title>How to run Claude Code in the cloud without patching it</title>
      <link>https://useagent.org/blog/run-claude-code-in-the-cloud</link>
      <guid isPermaLink="true">https://useagent.org/blog/run-claude-code-in-the-cloud</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Claude Code</category>
      <category>Gateway</category>
      <description>How we run Claude Code in the cloud unmodified and configured per run, why a 128K output default read as auth failures, and the 64,000-token fix that held.</description>
    </item>
    <item>
      <title>Self-hosted AI coding agent: what one host really needs</title>
      <link>https://useagent.org/blog/self-host-ai-coding-agent</link>
      <guid isPermaLink="true">https://useagent.org/blog/self-host-ai-coding-agent</guid>
      <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
      <dc:creator>Abhishek</dc:creator>
      <category>Engineering</category>
      <category>Self-hosting</category>
      <category>Postgres</category>
      <description>What it takes to run a self-hosted AI coding agent: one Linux host, Postgres with pgvector, one backend per database, a sandbox provider, mail, backups, disk.</description>
    </item>
  </channel>
</rss>
