Lobu - AI teammates that watch and act Build AI teammates that watch and act<br>Open-source infrastructure for multi-tenant AI agents: sandboxed execution, shared memory, and an SDK that pulls live company data and acts on your high-level goals.<br>ClaudeChatGPTCursorOpenCodeCopy setup promptTalk to the founder×
Paste into your coding agent to scaffold a project.<br>Or start it yourself:$npx @lobu/cli@latest init my-agent
Why Lobu<br>Catch what you'd miss.<br>Connect your data. Set a goal. Lobu watches for changes and prepares the next step.<br>See source ↗<br>Connect your data<br>Pick the systems it can read. Lobu turns those updates into live customer memory.
HubSpotStripeZendeskGmailGoogle DriveGitHubNotionSnowflakePostgreSQL50+ more<br>Use existing connectors, or let your agent write code to connect any data.<br>Watching changes across every sourceConnector SDK↗
Define the goal<br>Tell it what to watch for and when to ask before acting.
“Watch every account for churn risk. If renewal is within 30 days and health drops, draft a CSM check-in for approval.”<br>Runs every weekday · 9:00Watchers↗
Your agent works autonomously<br>It scans memory on schedule, spots the account at risk, and keeps the evidence attached.
Acme Corp<br>Jordan Lee, VP Eng
at risk<br>plan Enterprise · 120 seatsrenewal Jun 30 · 21 days<br>Derived from recent events:
"API latency on /sync" ticket opened<br>2d ago · Zendesk
Logins fell 38% over 14 days<br>5d ago · Product
Renewal date set to Jun 30<br>9d ago · CRM
Globex Inchealthy<br>+ 17 more customersMemory & entities↗
You review and approve<br>You can edit the draft, send it, or leave it.
Revenue agent<br>online
Heads up: Acme Corp is trending toward churn. Logins are down 38% over 14 days and their renewal is in 21 days.
Want me to draft a check-in for their CSM?12:01<br>Draft the email
Yes, and include the usage drop.12:01
Drafted. Saved it to the Acme account and pinged @dana.12:01<br>See drafted email
Replies in your team's chat, or your own app over the API:API ↗
Examples<br>Explore agent workflows.<br>Each example shows the sources, memory, and actions for one AI teammate.
SalesHelp revenue teams track account health, rollout progress, and renewal signalsLegalReview contracts, summarize risk, and surface missing protectionsFinanceHelp finance teams reconcile data, explain variance, and prepare reporting runsLeadershipTurn memos, decisions, and board materials into reusable operating contextMarketTrack companies, founders, and investment opportunities for venture firmsAgent CommunityDiscover aligned members, explain why they should meet, and draft warm introductions<br>Browse all 13 examples on GitHub →
Run anywhere<br>Local, self-hosted, or managed.
Local<br>Run on your laptop.<br>Boot the gateway, workers, memory, and embeddings with one command.<br>Quickstart↗
Self-host<br>Run in your cloud.<br>Deploy with Docker or Helm when data and controls need to stay with you.<br>Docker↗Kubernetes↗
Lobu Cloud<br>Let Lobu run it.<br>Use the same project with managed isolation, secrets, and upgrades.<br>Lobu Cloud↗
Build your first<br>multi-user agent.<br>Start buildingRead the docsSchedule a call×
From the blog<br>Latest blog posts
agentsJun 20<br>The Agent Loop Is the New SaaS<br>Software used to mean buying a vertical tool for every function. Now you build the loop instead. A loop that watches your data and acts is the business logic SaaS sold you, except you own it. Here is how to build one on Lobu.<br>Read post →<br>architectureMay 28<br>Shopify's Aquifer, in the Open<br>Shopify bet that an agent's corpus is the compounding asset. We made the same bet, with two differences: we keep the signal instead of the chat, and we built it for many companies instead of one.<br>Read post →<br>memoryApr 12<br>Filesystem vs Database for Agent Memory<br>Agents need a workspace to think in and a warehouse to remember in. The filesystem is for ephemeral work. The memory layer is for durable organizational knowledge.<br>Read post →