Code mode yields a 99.2% cost reduction in our systems • Hacker News Board

Re: Code mode yields a 99.2% cost reduction in our systems

#71

Post by tarasyarema » Thu, Jul 23, 2026, 1:53 PM UTC

vessenes wrote: post #66

tarasyarema wrote: post #32

Earlier quoted context omitted.

thanks!

How’s the agent swarm working? Do you have like a sample startup you run using it?

We run it ourselves since nov last year (~20k tasks since then, note we are 2 people and bootstrapped).

Then we know a few companies using it w +1k tasks x day (~20/30 people actively using it daily). Then we have some anonymous telemetry too, we are seeing more adoption, but as it's a toggle we only know about the companies that talk w us directly / contribute to the repo.

The type of startups, we generally see tech adopting initially due to the technical deployment required (e.g. k8s), but then tech managers, product, marketing and ops starting to use it more. Connecting integrations, and the Slack integration is super good, works as if it was a peer.

Re: Code mode yields a 99.2% cost reduction in our systems

#72

Post by sourcecodeplz » Thu, Jul 23, 2026, 2:05 PM UTC

Codex did this for me by itself when i asked it for some daily automations.

it just wrote scripts.. scripts are coming back

Re: Code mode yields a 99.2% cost reduction in our systems

#73

Post by josefritzishere » Thu, Jul 23, 2026, 2:42 PM UTC

I think ECREE applies. "Extraordinary claims require extraordinary evidence" Nothing is 99% effective. This reads like a marketing piece.

Re: Code mode yields a 99.2% cost reduction in our systems

#74

Post by killix » Thu, Jul 23, 2026, 10:55 PM UTC

shireboy wrote: post #5

A maxim I’ve arrived at working w llm every day is “put deterministic things in code, non deterministic in llm”. I do wish the harnesses would be more helpful in this regard. For example I’ve seen tons of scheduled jobs that people wrote in Claude/copilot/etc that could easily have been scripts. They aren’t scripts because the author doesn’t know how to script and they stop at the point the llm does what they want. I…

[flagged]

Re: Code mode yields a 99.2% cost reduction in our systems

#75

Post by rajeevbakshi » Fri, Jul 24, 2026, 5:28 AM UTC

moberemk wrote: post #23

tarasyarema wrote: post #16

Earlier quoted context omitted.

Hi, the blog post was written assisted with our own swarm, based on the experiment. Indeed we had the concept of workflows from the start (like n8n), but the point of the scripts is type safe executions that the agents can write, with access to APIs, MCPs, etc. E.g. this PR from ~10 days ago https://github.com/desplega-ai/agent-swarm/pull/934 was going in this direction, offering a way to defined type safe connection…

No offence but, “if you find yourself doing the same thing over and over again write a script for it” isn’t exactly novel thinking in the software field…this was the same process when it was a human doing it, we just measure that in time instead of tokens

[flagged]

Re: Code mode yields a 99.2% cost reduction in our systems

#76

Post by tarasyarema » Wed, Jul 29, 2026, 7:19 PM UTC

Hello all OG here. For anyone coming back to check the comments, thanks a lot for all of them and the feedback!

I did a re-write of the post to remove the sloppy parts and making more clear some of the estimates and assumptions we made in the post.

Also linked this thread in there, so that anyone getting to that post can see all the interesting pieces mentioned in here.

Till the next on HN!!1

Re: Code mode yields a 99.2% cost reduction in our systems

#77

Post by yencabulator » Thu, Jul 30, 2026, 7:18 PM UTC

firasd wrote: post #51

shireboy wrote: post #5

A maxim I’ve arrived at working w llm every day is “put deterministic things in code, non deterministic in llm”. I do wish the harnesses would be more helpful in this regard. For example I’ve seen tons of scheduled jobs that people wrote in Claude/copilot/etc that could easily have been scripts. They aren’t scripts because the author doesn’t know how to script and they stop at the point the llm does what they want. I…

An absurd example is how harnesses like Codex, Claude code have no 'cut/paste this range of chars' affordance so if you want to move a block of code the LLM galaxy brain rewrites it in token space in the new file. It ~can~ be done using standard bash utils but there just isn't a kind of standard pattern for 'move this text block via CLI tools' so the LLM and human both overlook it

Worse, the content subtly mutates during that transformation. It's really rare to see that happen with anything affecting execution, but I see comments lost/created all the time.