jared_stewart

6 ms·

1.
by jared_stewart 16d ago
Location: Denver, CO Remote: yes Role: Staff+ Engineer, Tech Lead, Engineering Manager I'm a serial builder who thrives in the space between engineering and business. Over the last 10 years I've: - Designed autonomous driving tech
2.
by jared_stewart 7mo ago
The server exposes a straightforward API so wrapping it in MCP should be straight forward. The agent / skill interacts with the server using the cli implementation (part of the skill definition) at https://github.com/Ja
3.
by jared_stewart 7mo ago
Typed languages definitely provide richer signal in there signatures - and my experience has been that I get more reliable generations from those languages. On the efficiency point, the agent isn't doing any expensive exploration here.
4.
by jared_stewart 7mo ago
I see, so the context adapts as the LLM interacts with the codebase across requests? That's a clever implicit flow for ranking. The difference in my approach is that exploration is happening within a single task, autonomously. The agen
5.
by jared_stewart 7mo ago
will take a look at opencode, thanks for sharing!
6.
by jared_stewart 7mo ago
Aider's repo-map concept is great! thanks for sharing, I'd not been aware of it. Using tree-sitter to give the LLM structural awareness is the right foundation IMO. The key difference is how that information gets to the model. Aid
7.
by jared_stewart 7mo ago
I've been tinkering with it substantially and the most I can say is that it generally doesn't trigger automatically :( Claude has a really, really strong affinity for it's existing tools for exploring a code base. I'd be
8.
by jared_stewart 7mo ago
I don't see why it wouldn't - but I'm not familiar with setup / integration on other platforms. Would love to hear more about your stack and see if we can't find a way for you to try it out
9.
by jared_stewart 7mo ago
Tree-sitter and LSP solve different problems. LSP is a full fledged semantics solution providing go-to-definition functionality, trace references, type info etc. But requires a full language server, project configuration, and often a workin
10.
by jared_stewart 7mo ago
anecdotally, it seems like this helps find better places for code to sit, understands the nuances of a code base better, and does a better job avoiding duplicate functionality. it's still very much a work in progress, the thing I'
11.
Show HN: CodeRLM – Tree-sitter-backed code indexing for LLM agents(github.com)
81 points by jared_stewart 7mo ago|37 comments
12.
by jared_stewart 11mo ago
I've been working to build some tools for detecting and monitoring lookalike domains - the kinds of things used in phishing / brand impersonation attacks. My current prototype scans potential lookalikes for a target domain and the
13.
by jared_stewart 1y ago
Location: Denver, CO Remote: yes Role: Staff+ Engineer, Tech Lead, Engineering Manager, or Co-founder I'm a serial builder who thrives in the space between engineering and business. Over the last 10 years I've: - Designed autonomo
14.
by jared_stewart 1y ago
Location: Denver, CO Remote: yes Role: Staff+ Engineer, Tech Lead, Engineering Manager, or Co-founder I'm a serial builder who thrives in the space between engineering and business. Over the last 10 years I've: - Designed autonomo