Show HN: I built an interactive map of jobs at top AI companies

Show HN: Ami – A local, open-source agent that does your busywork across apps

Show HN: Ami – A local, open-source agent that does your busywork across apps Hey everybody, sharing Ami on HN today.<p>Ami is an open source, local-first agent harness that acts as your shadow worker and copilot chat. It ships with a graph memory.<p>Here&#x27;s what Ami does on its own -<p>- connects to apps, data, repositories, tools with your personal tokens<p>- Learns how you do tasks (execution style, decisions, anti-patterns)<p>- Learns how you communicate (external and internal)<p>- maintains a universal to-do list<p>Here&#x27;s how you use Ami -<p>1. You can execute busywork. It fetches and executes tasks autonomously in your style, asks approval before risky actions, gives deliverables, drafts replies &#x2F; emails &#x2F; ticket updates.<p>2. You can execute copilot chats. Use it to ask questions, fire off ad-hoc tasks, create to-dos, update memory.<p>Ami was built for internal use. My team found it useful, so we wanted to share it here. It&#x27;s still in development stage, and we might push a more stable release soon. It constructs a context graph memory of you, with entities, relationships, feedbacks, decisions, writing styles maintained in memory so it can get more autonomous the more you use it.<p>Few examples where Ami helped me this week -<p>1. fetched a bug report from slack, created fix PR autonomously which I merged, verified fix is working.<p>2. debugged a traffic spike on our new blog.<p>3. turned a sales POC into an order form draft using recently signed forms.<p>4. nailed down metrics definitions from notion and created a dashboard.<p>5. closed out my day by auto-updating Linear tickets based on slack activity.<p>Any feedback is most welcome.