pls-dnt-deploy

It seems to me as a management problem, not a tech problem.

You can delegate responsibility, not accountability.

This is a key principle in management, and usually what makes people progress in their career: if you take accountability for the failures, you also own the successes.

As I see it, if someone is pushing accountability down the chain, they are not the leaders you want.

check -> agent-swarm.dev (OSS/MIT) / repo: https://github.com/desplega-ai/agent-swarm

It's exactly that.

The key point is -> dynamic workflows, or static ones should contain guardrails and scripts enforcing certain rules. Check the code!

It's non-trivial for sure

How are you handling potential prompt injections when, for example, you are browsing the web?

Are you limiting access for pre-defined activities based on the rules of a particular task?

I've been leaning towards investing more time writing, to balance all the random reading.

Also, trying to put more and more focus time, in which I don't allow myself to use AI.

I find it very rewarding.

There's a plot twist, you need to read it.

But if your lead/workflow is powered by one of those, you are ducked, right? How do you solve for that? Or is the retry/routing deterministic?

I would love to know why wouldn't you just trust devin? It's easy to swap regardless and pricing doesn't seem that bad.

Hi all,

With the change in policy from Github stargazers API, now it's not possible to do this at scale anymore. Instead, we did it with our own data, trying to discover what are those new repos stargazers we trust are following now.

If you are asking for the winners: * tracewayapp/traceway (1.2k ) Highest cohort penetration. OpenTelemetry for agents.

* shulcz/deja-vu (500 ) Couple of weeks old, memory retrieval system.

These are the results. Stay curious,

That loneliness reminds me of a creative process looking for a blackswan. Years, <building/creating/innovating> by yourself, trying to give value and deliver something that matters.

Trust starts to be key in those scenarios, and LLMs definitely don't have it from us rn.

Great article, I loved it! A few parts will stay with me.

The Idiosyncrasy dimension seems to be what I was missing the most. That relative and social aspect beyond discernment and pattern recognition.

This is the powerful quote, because taste wasn't a top concern until mass production appeared: "Tech’s surging obsession with taste is therefore not really about aesthetics; it reflects a deep anxiety about whether economic capital can capture cultural capital at the exact moment cultural capital becomes crucial for technology’s distribution, relevance, and differentiation."

Thanks!!!

I'm seeing the same thing, probably all will land in a combination of Antigravity Inbox experience, Devin and some OpenClaw omnipresence style UX.