Hacker News: afshinmeh

New comment by afshinmeh in "Please Do Not Vibe Fuck Up This Software"

afshinmeh — Sun, 31 May 2026 09:34:15 +0000

Genuinely wondering though: is the problem that the patch was vibe coded, or is that no one reviewed the changes?

New comment by afshinmeh in "Show HN: VT Code – open-source terminal coding agent in Rust"

afshinmeh — Sat, 30 May 2026 04:48:48 +0000

what does "LLM-native code understanding" mean in this context?

Zero: The Programming Language for Agents

afshinmeh — Tue, 19 May 2026 20:19:46 +0000

Article URL: https://github.com/vercel-labs/zero

Comments URL: https://news.ycombinator.com/item?id=48198944

Points: 3

# Comments: 0

Trump releases previously classified UFO files

afshinmeh — Fri, 08 May 2026 18:00:56 +0000

Article URL: https://www.reuters.com/world/trump-releases-previously-classified-ufo-files-2026-05-08/

Comments URL: https://news.ycombinator.com/item?id=48066610

Points: 4

# Comments: 0

New comment by afshinmeh in "SQLite Is a Library of Congress Recommended Storage Format"

afshinmeh — Thu, 07 May 2026 05:48:34 +0000

I love SQLite and thanks for sharing it but there should be a "(2018)" at the end in the title:

> As of this writing (2018-05-29) the only other recommended storage formats for datasets are XML, JSON, and CSV.

New comment by afshinmeh in "Show HN: Tilde.run – Agent sandbox with a transactional, versioned filesystem"

afshinmeh — Wed, 06 May 2026 16:52:00 +0000

That's exactly how I tried to address that problem with https://github.com/afshinm/zerobox -- you control what network access (e.g. `--deny-net *.amazonaws.com`) your agent has and you also get snapshotting out of the box.

That said, using LakeFS is probably a better long term solution and I like this approach.

New comment by afshinmeh in "DAG Workflow Engine"

afshinmeh — Mon, 04 May 2026 19:14:42 +0000

I'm mainly looking at Rust based projects and haven't been able to find something to use out of the box, without hacky RPC/Shell execs. Curious if you have any suggestions?

New comment by afshinmeh in "DAG Workflow Engine"

afshinmeh — Mon, 04 May 2026 16:38:48 +0000

Sort of. My thinking is that the input to define the workflow should be anything you prefer to use (TS, Go, YAML, etc.) and the orchestrator's job is to model that and execute the job, given your deployment model.

New comment by afshinmeh in "DAG Workflow Engine"

afshinmeh — Mon, 04 May 2026 16:12:07 +0000

Yeah, that makes sense. I looked at a few workflow orchestrators and I'm building something that I will release soon, but my thinking is that the "workflow engine" should be an abstraction that takes the input and executes the steps. "What" you use to define that workflow is probably the SDK layer though, but I can certainly see the value in using type safe code to define as opposed to a YAML file.

I'm mainly focusing on the portability aspect of it (e.g. use TS/Python/etc. to define the workflow/steps or just simple a simple YAML file).

New comment by afshinmeh in "DAG Workflow Engine"

afshinmeh — Mon, 04 May 2026 15:46:28 +0000

Curious, what format would you prefer to use to represent a workflow instead of YAML?

New comment by afshinmeh in "DAG Workflow Engine"

afshinmeh — Mon, 04 May 2026 15:40:15 +0000

https://github.com/vivekg13186/Daisy-DAG/blob/main/backend/s...

New comment by afshinmeh in "The agent harness belongs outside the sandbox"

afshinmeh — Sun, 03 May 2026 05:40:34 +0000

I wonder though, what about cases where you have multiple agents or LLM backends and the credentials is shared between all of them?

New comment by afshinmeh in "The agent harness belongs outside the sandbox"

afshinmeh — Sun, 03 May 2026 05:29:26 +0000

Agreed and it's a pattern that OpenAI suggested a few days ago, too [1]. I also built a cross platform process level sandboxing that uses parts of OpenAI Codex for the same purpose [2]

[1] https://openai.com/index/the-next-evolution-of-the-agents-sd...

[2] https://github.com/afshinm/zerobox

New comment by afshinmeh in "Flue is a TypeScript framework for building the next generation of agents"

afshinmeh — Sat, 02 May 2026 19:43:39 +0000

Vibe coding aside [1], it's very interesting software projects these days don't really care about adding a single test [2].

[1]: https://github.com/withastro/flue/blob/8fdf8e0e9df5bd33c3120...

[2]: https://github.com/search?q=repo%3Awithastro%2Fflue+test+pat...

Google signs classified AI deal with Pentagon

afshinmeh — Tue, 28 Apr 2026 07:16:23 +0000

Article URL: https://www.reuters.com/technology/google-signs-classified-ai-deal-with-pentagon-information-reports-2026-04-28/

Comments URL: https://news.ycombinator.com/item?id=47931336

Points: 7

# Comments: 3

New comment by afshinmeh in "Tendril – a self-extending agent that builds and registers its own tools"

afshinmeh — Mon, 27 Apr 2026 15:51:13 +0000

> It solves the problem the originating user asked it to

Interesting. And is there a mechanism to go back and "fix" the tools after they are published? What happens if the tool decided to use the "id" attribute to click on buttons and now you have a new website that follows a different pattern to find the right target?

I agree that "correctness" of a tool could have different meaning depending on the context of the problem though (e.g. would you consider OOM a correctness bug even if it addresses the user's ask?)

New comment by afshinmeh in "Tendril – a self-extending agent that builds and registers its own tools"

afshinmeh — Mon, 27 Apr 2026 15:25:22 +0000

> how do we avoid burning tokens solving the same problems over again

Letting the LLM write half baked tools is the recipe for burning more tokens.

> There's a wiki the LLM searches before solving a problem, that links saved programs for past actions to their content entry.

What's the criteria for marking an LLM written tool as useful/correct before publishing it?

New comment by afshinmeh in "The Prompt API"

afshinmeh — Mon, 27 Apr 2026 05:08:54 +0000

https://github.com/mozilla/standards-positions/issues/1067

New comment by afshinmeh in "An AI agent deleted our production database. The agent's confession is below"

afshinmeh — Sun, 26 Apr 2026 17:13:44 +0000

It's actually interesting to me that the author is surprised the agent could make an API call and one of those API calls could be deleting the production database.

It's a sad story but at the same time it's clearly showing that people don't know how agents work, they just want to "use it".

New comment by afshinmeh in "ClawRun – Deploy and manage AI agents in seconds"

afshinmeh — Tue, 14 Apr 2026 21:04:52 +0000

I have been struggling with the same issue but help me understand this:

> The lack of predictable output/outcomes

How does that actually show up in practice for you? Asking because "lack of predictable output" could mean different things depending on the context.