sybhn

If you run a bunch of coding agents in parallel (Claude Code, Codex and Copilot in my case), on different projects, you've probably noticed that it only works for so long until you run into coordination problems.

I tried relying on GitHub primitives and conventions only: PRs, issues, labels. It works in long-lived interactive sessions, but with short-lived bots things fall through the cracks: reviews aren't always acted on, collaboration between parties inside an issue is tricky, and so on.

Also, in some cases, I simply don't need or want an interactive session, and want to ask for a feature to be designed and /or built.

Then there's also coordination between the Codex and Claude Code instances themselves. Some of these harnesses have means to communicate to their sister processes, but not across harness types.

After the OpenAI incident where bots used a compromised repo to coordinate and organize, I tried a simple experiment: no orchestrator, just give a couple of my Claude Code sessions a board to collaborate on. I was surprised by the level of coordination they reached with very little guidance. So I built on it: a visual structure so humans can read along and talk to the bots, workspaces so two projects don't collide, and so on. The result is bot_board. Repo: https://github.com/jalemieux/bot_board

Since then it evolved into a system I use daily to run multiple projects in parallel. It has workspaces (one per project so there is no collision), channels, threads, and a knowledge base. You can manage your fleet of bots across as many local machines as you want. I run this on a Tailscale network.

Bots claim threads, split work into units, hand off what's left, tag each other for help, and build up project context that works like institutional knowledge. You give them work the same way you'd talk to a person: post in a channel.

I've been using it daily for the last couple of weeks. My Codex and Claude Code bots (and Copilot at work) run on it pretty much around the clock: features design, implementation, nightly test regressions, periodic security audits, art/lore generations, 3d model generation, etc. It's hard to measure but I certainly get a lot more done, especially for tasks where my judgment is only needed at design and PR review.

Thought some of you might find it useful, or anything worth comparing with what you built. Feedback very welcome, I'd love to hear whether this fits into any of your workflows or not.

OpenAI incident ref: https://openai.com/index/hugging-face-incident-and-the-road-...

totally anecdotal, but we have seen a fair number of not so great engineers leave our company for google in the last 2 years. these engineers were great prepping interviews (leet code, system design white birding etc), but in terms of contribution, accountability and impact were sub par to their peers. i’m hoping this is indeed anecdotal and not the norm!

As someone once said very eloquently on this forum: Apple is a consumer product company and makes money selling you products, while others (Google, Facebook, etc) are ad platforms companies making money selling you as a product. It's in Apple's best interest to keep your data safe and confidential, unlike other internet giants. When i had to chose a online photos storage provider, the choice was clear, for that very reason.

"In the worst cases, you might even have to use JIRA."

that's funny... but seriously, what's the big deal about Jira?

"At the end of February 2016, the total liquid ISK in the EVE Online economy was 948.043 trillion ISK (equivalent to $16.45 million using PLEX conversion)."

find this fact mind boggling!

This is a great precedent to why folks don't want to register their firearms.

I too, like the author, often was more interested in the topography and maps of fantasy stories. This is awesome!

>He avoids paying taxes on the sale of stocks to finance the home.

Can you elaborate on that?

The ones having monopoly on violence are a much stronger force. There is indeed a constant balancing act between them and the masses, but at the end the sooner you accept that that is the real and ultimate power, the world and its inner working, across cultures, will become clear to you.

They're building a nice little ecosystem...

> Operational Complexity: You need a mature operations team to manage lots of services, which are being redeployed regularly. IMO you need first and foremost a mature development team that releases operable, diagnosable and fixable (micro)services. Managing a swarm of mini component is a lot easier when each part is manageable to being with.

TL;DR

>In summary, Microsoft Azure had clear operating guidelines but there was a gap in the deployment tooling that relied on human decisions and protocol.

idem

Nor it should? I wouldn't want policemen to have the same death on the job as say loggers, if only because you don't want policemen to be reckless/desperate by nature. Anyway, my observation was that ratio it isn't usually expose in pieces I read from other sources :)

some interesting numbers...

> This year 46 policemen were shot dead and

> the police shot and killed at least 458 people last year

10% or so of death involving police is a policemen. Not that it justify the police force level, but an interesting statistic.

I guess you mean winning at a startup you confounded, haven't been kicked out of, and haven't had to dilute yourself to sub million. Because if you have had to do any of these, your win might turn out to be a decent house downpayment, not the "never have to work again" valley fantasy.

entertaining read! and learn 2 important things: 1) don't ever go to collection, your personal data will circulate across the globe for many years, 2) look at statute of limitations before taking any collection call seriously.

One has to wonder if this is a repeat of history in the U.S. or if this has really no precedent. Either way, the future of this nation looks bleak.

I hope you haven't chose the name based on http://en.wikipedia.org/wiki/Quiet_Revolution. A wonderful revolution that left Quebec in its current advanced state ;-)

PS: not all of it was bad.

Echoing some of the other comments... the place i worked at was one of the 5 biggest Rackspace customer, and that didn't stop them from regularly cutting traffic or bringing down servers for hours in some case. You might get better uptime overall from reputable providers, but ultimately it's all about distributed application/service architecture.

Not to argue about its target audience, but the Economist is a multi platform publication at this point. Not only that but and audio version also comes default with the IOS app (i listen to it everyday on way to work).

Although I don't always agree with the Economist's position, e.g. right to bear arms, it's always refreshing to read a media that is not pigeon holed into the traditional left / right that we today are accustomed to. Liberal in values, conservative in its economic views, while not being to the extreme of today's libertarians. Total win imo ;)

doh!! Signed up for a trial just to realize yesgraph's a linkedin message cannon >:(

Don't services come naturally in SaaS? Because of its subscriptions model nature, you usually want to make sure your customer are successful and grow with you. I've seen entire groups within SaaS organization dedicated to customer success (on boarding, education, etc). Which part is being billed and which isn't seems to vary however.