Viewing profile — jswny
jswny
HN member- Joined
- Fri, Oct 09, 2015, 5:24 AM UTC
- HN karma
- 1,510
- Public activity
- 350 items
- HN profile
- View on Hacker News ↗
About jswny
[ my public key: https://keybase.io/jswny; my proof: https://keybase.io/jswny/sigs/pUFgX77M84hK-jJjciHEYOD7rL-iywoxAtKO-gALkT8 ]
Recent public activity
-
comment
Comment #49010291
Does anyone do TLA style distributed systems verification with Lean? Curious the experience there and how well supported it is
-
comment
Comment #48957908
Unfortunately I’ve used Claude and GPT models for a long time in a variety of harnesses and I agree with you and I think it’s the compaction. For some reason, codex compaction is l…
-
comment
Comment #48890947
I agree with some of Kelley’s takes, but the issue is the tone. Does anyone think that if Bun had been rewritten from Rust to Zig that a member of the Rust core team would have wri…
-
comment
Comment #48789232
What’s the advantage of that for building?
-
comment
Comment #48165152
Why not WASM?
-
comment
Comment #47881006
Codex is fully open source…
-
comment
Comment #47801875
You want to use the model that is potentially giving your data to the government vs the one that’s openly rejecting that partnership?
-
comment
Comment #47383564
Package a skill with your CLI itself and give users instructions on how to install the skill properly. That allows the agent to read the instructions in a context efficient way whe…
-
comment
Comment #47383555
Plenty of MCPs require you to install and run them locally, like I said remote MCP has a real advantage over CLI tho
-
comment
Comment #47381251
MCP loads all tools immediately. CLI does not because it’s not auto exposed to the agent, got have more control of how the context of which tools exist, and how to deliver that con…
-
comment
Comment #47381210
MCP is fine, particular remote MCP which is the lowest friction way to get access to some hosted service with auth handled for you. However, MCP is context bloat and not very good …
-
comment
Comment #47083986
Can you explain what’s so different about pro? I’ve used everything frontier model and had Pro a while ago but it seemed to just be the same models served faster at the time.
-
comment
Comment #46881363
How do you get sub agents to work?
-
comment
Comment #46744737
That’s just spawning multiple parallel explore agents instructed to look at different things, and then compiling results That’s a pretty basic functionality in Claude code
-
comment
Comment #46743282
What’s the point of having a public GitHub repo with PRs enabled if they will never merge any of them?
-
comment
Comment #46743235
How do you know/toggle which API path you are using?
-
comment
Comment #46740238
How does this work for other models that aren’t OpenAI models
-
comment
Comment #46449003
Ah I think it’s because I was testing with an external library in the top level TOML metadata
-
comment
Comment #46440807
The problem is that only Deno can type check single file scripts. Otherwise with Node and Bun you need a project to use tsc. Python can type check single file scripts (even with PE…
-
comment
Comment #46440795
I wish this was faster, in my tests it’s about 200ms startup consistently on my M4 MacBook. Otherwise very cool
-
comment
Comment #46440644
Bun has a lot of this built in, plus Bun shell
-
comment
Comment #46428904
Not memory, I just instruct it to freely experiment with temporary scripts and artifacts in a specific folder. This helps it organize temporary things it does like debugging script…
-
comment
Comment #46428327
Can you give an example of how beads would be used by Claude to do something it otherwise couldn’t? I can’t quite tell what it is useful for
-
comment
Comment #46428299
I also specifically instruct Claude how to use a globally git ignored scratch folder “tmp” in each repo. Curious what your approach is
-
comment
Comment #46365487
Very interesting project! I have 2 questions: 1. How has it been working with ACP? Is it anywhere near feature parity with Claude code’s native interface? 2. I see your repo is wri…