Brian Conn (@TheConnMan) header
Brian Conn (@TheConnMan) profile picture

Building software and teams with agents and humans. Founder of Connsulting, co-host of @managerxbuilder

Washington, DC
Joined May 2013
Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)Photo by Brian Conn (@TheConnMan)
Kunal defines burnout as aspiration versus expectation. If you choose to finish a project in two days because you love it, you're happy. When someone else expects that pace, that's burnout.
2
24
Once Brian splintered his workers, logging into MCP on every box became a pain. Now they all hit one mcporter gateway from OpenClaw, and the CLI works whether the harness wants config.toml or .mcp.json. managerxbuilder.substack.com…
1
55
Brian has broken his home lab more than once while setting it up. He keeps a vanilla fail-safe instance on Tailscale so Claude can still reach Proxmox and bring everything back online.
1
25
If you can reproduce the ticket, get a PR, and ship the fix without much human time, bugs get cheap. Cheap enough that you might choose to ship more of them so you can move faster. managerxbuilder.substack.com…
1
23
Stock DeepSeek refused to write the phishing email. Brian switched to the abliterated twin of the same model and it wrote the lure, reset link included. managerxbuilder.substack.com…
1
119
Klaus thinks dedicated tech debt weeks should be gone. Kunal's not sure, because you're shipping a lot more than you used to.
1
35
An abliterated model has the refusal weights cut out. Brian ran one on an isolated home lab next to stock DeepSeek. Same model family. One writes the phishing email, the other won't.
2
1
134
Brian deleted all of his custom sub-agents this week. He still runs vanilla sub-agents constantly, because a test writer shouldn't know what the implementer did. managerxbuilder.substack.com…
1
62
Brian's personal wiki ranks sources by trust. His own words count most, then his calls, then the podcasts and newsletters he trusts. Random web results never get in.
1
55
Tomorrow on Manager Builder, Kunal and Brian debate generalized vs specialized agents. "Most of our best practices are just best practices for humans."
1
60
In planning, Claude should ask what you want to do and then stop. Once it hands you a recommended option, it's too easy to just hit Enter. managerxbuilder.substack.com…
1
65
One-shot vs progressive on the same monitoring stream. One-shot: four false pages, missed the leak, a couple of cents. Progressive: one page, no false positives, about 50 cents. managerxbuilder.substack.com…
1
47
Asked a fast classifier to make one call per event. Four false alarms, and it missed the memory leak.
1
48
AI slop can cost you a client. Manager: keep a quality gate when models write the code. managerxbuilder.substack.com…
2
36
Why personal software matters: remove friction in your own workflow, and get a low-risk lab to try AI before work. Builder. managerxbuilder.substack.com…
1
37
Squarespace had no API, so I was the human in the loop adding zero value. One day-long side project with AI later, the site runs on Astro 5 + Cloudflare Pages. Builder take on personalized software.
1
1
1
84
Are you leading Claude, or is it leading you? Manager take: lead the bots instead of being led by them. managerxbuilder.substack.com…
2
1
35
Brian Conn retweeted
Cool eval. Simply ask an LLM “Land or Water?” and give it a latitude and longitude coordinate as text. Ask 16,200 times, plot as image. The models know. From compressing the internet.
Replying to @celestepoasts
results for all claudes
729
1,101
184
19,608
1,641,531
A human avatar sending dehumanized Claude-speak in Slack isn't helpful — it feels disrespectful. Manager take on AI slop in human channels.
1
1
46
Functional constraints vs architectural ones — which actually breaks your agent workflow?
2
1
32