We Used AI Agents for 7 Months. Here’s What Actually Happened
🚀 Take control of your online privacy! Try GoLogin for FREE: https://bit.ly/gologin_hause-collective
Promo code: HAUSE40 (The trial runs for 7 days, and the code gets you 40% off).
We’ve spent seven months actually using AI agents in our work, including OpenClaw and Claude Code. They’re making us capable of doing far more - but we’re increasingly wondering whether that’s making us more productive or simply giving us more work to do.
Anders runs one developer agent on OpenClaw with a Telegram topic per project. Yeeling runs ten named agents across her personal life and the business, with Claude Code for the daily work. A lot gets done while we sleep. The question in this episode is whether that has made us more productive, or just busier.
We go through what an AI agent actually is, how each of us set ours up and why the setups look nothing alike, feeding meeting transcripts straight to an agent, the interfaces that make it feel like work is happening, whether we still check the output, the weeks that went into setup, and what we would tell someone starting now.
Chapters
0:00 Are we more productive or just busier?
0:21 Seven months of running Hause Collective on AI agents
1:21 What an AI agent actually is
2:52 2026, the year of personal AI agents
3:51 A Mac mini and OpenClaw: how Anders set up
5:07 What OpenClaw is, and why it is risky
7:00 Claude Code, and why nothing runs on our laptops
9:09 Sponsor: GoLogin
10:34 Yeeling's setup: OpenClaw for automation, Claude Code for work
11:28 One developer agent, one Telegram topic per project
13:42 Feeding meeting transcripts straight to the agent
15:53 One agent that deploys everything on its own
17:28 How much has changed in seven months
18:23 Prompting from a phone, and the case for voice mode
21:16 Ten agents with names, and the weeks of setup
23:45 So, are we busier or more productive?
24:28 AI slop, and projects that never launch
26:34 The interface that makes it feel like work is happening
27:35 Are we still checking the output?
30:12 Trusting AI without noticing, and why evals matter
31:21 Agent Studio: one agent tests, another one fixes
34:32 Has life actually gotten easier?
35:29 The setup takes time: false productivity and training an intern
36:29 First movers now, mainstream next year
37:04 Just get started
38:58 Wrap up
39:23 Outro
If you are running agents, are you more productive or just busier? Tell us in the comments, we read and reply to all of them.
This video is sponsored by GoLogin.
Music
"Retro Energy (30s)" by Universfield (https://freesound.org/people/Universf...) — CC BY 4.0
"Beauty Flow" Kevin MacLeod (incompetech.com) Licensed under Creative Commons: By Attribution 4.0 License http://creativecommons.org/licenses/b...
Follow Hause Collective:
X: @HauseApp
IG: @HauseCollective
LinkedIn: /hause
Website: https://hause.co/
#AIAgents #OpenClaw #ClaudeCode