kenforthewin

6 ms·

61.
by kenforthewin 8mo ago
If I understand correctly, the vecstore product is an api that wraps a vector database (among other features). So the pitch here is something like “you don’t need a vector database, you need an api wrapper to a vector database that we manag
62.
by kenforthewin 8mo ago
I'm glad it's those kinds of mines rather than the ones I first thought of.
63.
by kenforthewin 8mo ago
Does an AI also playing your game somehow detract from the pleasure you derive from it? I find it entertaining both to play the games, and see how LLMs perform on them; I don't see how these are in any way mutually exclusive.
64.
by kenforthewin 8mo ago
Nice work. I think games are a great way to benchmark AI, especially games that involve long term strategy. I recently built an agent harness for NetHack - https://glyphbox.app/ - like you I suspect that there's a lot
65.
by kenforthewin 8mo ago
If that's true, it sounds like the vibe coders are winning - they're creating products people want, and pull in technical folks as needed to scale.
66.
by kenforthewin 8mo ago
Hey HN - I built an agent harness for NetHack that exposes a Python sandbox for agents to write game commands and script their way to ascension. More recently I built a web app around this framework that allows anyone to watch the agents pl
67.
Show HN: NetHack agent harness with benchmarks and livestream(glyphbox.app)
1 points by kenforthewin 8mo ago|1 comments
68.
by kenforthewin 9mo ago
We've been optimizing for decades to engineer the bullshit-generating super-soldiers required to craft modern PR statements.
69.
by kenforthewin 9mo ago
...what?
70.
Show HN: Watch AI Agents Play NetHack(glyphbox.app)
1 points by kenforthewin 9mo ago|2 comments
71.
by kenforthewin 9mo ago
Hey HN - I built an agent harness for NetHack that exposes a Python sandbox for agents to write game commands and script their way to ascension. More recently I built a web app around this framework that allows anyone to watch the agents pl
72.
Show HN: Watch LLMs Play NetHack(glyphbox.app)
1 points by kenforthewin 9mo ago|1 comments
73.
by kenforthewin 9mo ago
Let's add NetHack to the mix! https://kenforthewin.github.io/blog/posts/nethack-agent/
74.
It's 2026. Can LLMs Play Nethack Yet?(kenforthewin.github.io)
2 points by kenforthewin 9mo ago|0 comments
75.
It's 2026. Can LLMs Play Nethack Yet?(kenforthewin.github.io)
1 points by kenforthewin 9mo ago|0 comments
76.
by kenforthewin 9mo ago
> This is insanity masquerading as pragmatism. > This is not engineering. This is fashion masquerading as technical judgment. The boring explanation is that AI wrote this. The more interesting theory is that folks are beginning to ado
77.
by kenforthewin 10mo ago
Just a reminder that everyone copes with change differently.
78.
by kenforthewin 10mo ago
I find the cost discussion to be exceedingly more tedious. This would be a more compelling line of thinking if we didn't have highly effective open-weight models like qwen3-coder, glm 4.7 etc. which allow us to directly measure the cos
79.
by kenforthewin 11mo ago
Sure, but Augment’s main value add is their context engine, and imo they do it really well. If all they had to do was launch an MCP for their context engine product to compete, I think the comparison is still worth exploring.
80.
by kenforthewin 11mo ago
Congrats. From my experience, Augment ( https://augmentcode.com ) is best in class for AI code context. How does this compare?
81.
by kenforthewin 11mo ago
It's fascinating to see the evolution of HN sentiment towards LLMs in real time. Just a few months ago, projects like these were a dime a dozen and every AI-related post had a skeptical comment at the top. Now I'm almost surprised
82.
by kenforthewin 11mo ago
Similar project: https://github.com/aberemia24/code-executor-MCP And the original Anthropic post that inspired both: https://www.anthropic.com/engineering/code-execution-with-mc...
83.
by kenforthewin 11mo ago
No mention of coding benchmarks. I guess they've given up on competing with Claude and GPT-5 there. (and from my initial testing of grok 4.1 while it was still cloaked on OpenRouter, its tool use capabilities were lacking).
84.
Provider Variance: Introducing Exacto(openrouter.ai)
2 points by kenforthewin 11mo ago|0 comments
85.
by kenforthewin 11mo ago
If, like me, you were wondering what PDS stands for: https://github.com/bluesky-social/pds
86.
Show HN: OpEx, an agentic LLM toolkit for Elixir(github.com)
4 points by kenforthewin 11mo ago|0 comments
87.
by kenforthewin 11mo ago
GitHub link 404s
88.
Self-Directed AI Dev: Cellular Automata(kenforthewin.github.io)
1 points by kenforthewin 11mo ago|0 comments
89.
Experiments in Autonomous AI Development(kenforthewin.github.io)
1 points by kenforthewin 11mo ago|0 comments
90.
by kenforthewin 11mo ago
An AI development workflow platform with GitHub integration. Built in Elixir / Phoenix. Early stages but it's a fun project. https://github.com/kenforthewin/matic

More ›