cal (@calummoore) header
cal (@calummoore) profile picture

cto, yc w20 @conductivebuild

UK
Joined March 2009
Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)Photo by cal (@calummoore)
two masters trap developers - want to see things that need action, and switch quickly between them product managers - want to see an overview (more like this) product ends up serving neither
if you ADE doesn't do this, you ngmi
1
4
215
if you use a cloud sandbox for every workspace, then you don't need to worry about memory usage of sub-processes
Guys, plz don't let this become an IQ test.. These memory usage comes from AGENTS running inside of Ghostty or Orca. (e.g. a single Claude Code / Next.js server is a few GB already) We help you run hundreds of agents / dev servers. And these things cost memory! Not Ghostty or Orca itself.
2
246
something Conductor could never do
4
381
I have a skill that forces the agent to give me proper context
1
2
337
how I like to organise my work
1
5
651
who is building the alternative to github
5
2
7
874
Fine, if you have unlimited AI budget
This is f*cking insane. Anthropic engineer: "You're not supposed to prompt Opus 5.5 You're supposed to build a system that prompts itself. At Anthropic, 90% of our engineers have already built these systems" In 30 minutes, she walks through exactly how Anthropic builds agents that remember, fix their own mistakes and get sharper with every run. Better than any $1000-$2000 agent course I've seen. Watch it, then read the article below ↓
Readers added context they thought people might want to know
Lamis never says "You're not supposed to prompt Opus 5.5” in here talk at AI Dev Con. The caption misrepresents the video knowing most won’t watch the video all the way through and jump to the articles. youtube.com/watch?v=tTcxVv…
4
961
💯 saas isn't dead, simple saas is dead
If building is easier, build harder things.
1
3
472
founder p**n
everyone at @firecrawl runs their own team of AI agents +100 agents (captains) across the company, each with its own subagents and they all talk to each other context 24/7
1
1
378
Opus 5.5 benchmark just landed
2
681
how I use my agents planning -> fable frontend -> opus 5 max thinking backend -> codex 5.6 max thinking Fable orchestrates the other agents - I just prompt it with the above split. Usage goes pretty far with this.
1
3
721
this is getting out of control
I think I just cooked something 🔥 jev(): a PostgreSQL extension that searches your whole database in natural language. No index, no embeddings, just one function. WHERE jev(people, 'could work from home') or WHERE jev(people, 'name sounds european') 129 rows judged in ~1s for $0.0009. Second run: 6ms from cache.
1
750
very dangerous prompt
5
527
we spent years removing latency from software now I wait 20 minutes for an agent to confidently do the wrong thing
2
4
593
this is what my job has been reduced to
4
541
burned $600 because my agent found an api key I was using for testing and decided that was better than my subscription 😭
6
40
2,941
I hope someone is working on the next github no new features, just something that works
4
3
580
people need to be more honest
1
3
489
this is actually not bad given the basic prompt: "create the entire rollcoaster tycoon in the web browser"
1
3
512