austinbaggio

8 ms·

31.
Show HN: 20+ Claude Code agents coordinating on real work (open source)(github.com)
53 points by austinbaggio 8mo ago|39 comments
32.
by austinbaggio 9mo ago
SWE moves up the stack. Juniors need to get really really good at shipping demos quickly, and writing excellent requirements based on core user need. ack bias, former PM
33.
Show HN: Why single agents suck at math proofs(ensue.dev)
4 points by austinbaggio 9mo ago|1 comments
34.
by austinbaggio 9mo ago
Willing to let them loose is the more salient point. If you let your agents loose on your entire body of output and tools at work, then you'll build that knowledge up pretty quickly. Tall ask right now, with privacy and agency (no pun
35.
by austinbaggio 9mo ago
Same, I was a very average dev coming out of CS, and a PM before this. I find that my product training has been more useful, especially with prototypes, but I do leave nearly all of the hard system, infra, and backend work to my much much m
36.
by austinbaggio 9mo ago
| need a solution for context decay and relevant curation — with benchmarks that prove it is also more valuable than constant rediscovery (for quality and cost). I agree. We are looking at some metr benchmarks, not expecting a simple answer
37.
by austinbaggio 9mo ago
What are you both looking for? What is the problem you want solved?
38.
by austinbaggio 9mo ago
Oh got it - tool use
39.
by austinbaggio 9mo ago
If you give it a try, I think that use case should work, but if not, I would be grateful if you told us what broke. And also - I genuinely worry about vendor lock-in, do you?
40.
by austinbaggio 9mo ago
I can't find it, but there's a good graph that shows Google search decline in share to GPT, but it excludes Gemini. With Gemini, it stays relatively on par. That's pretty much the answer with where one goes. LLMs are higher i
41.
by austinbaggio 9mo ago
Yeah I get the dependency concern, and also I think about the trust and pricing challenge a lot. I might be getting ahead of my skis here, but living in a future world, assuming there is a local service, what would you want to see with a co
42.
by austinbaggio 9mo ago
Thanks everyone for the comments, really, I wasn't expecting this. Quite a few of you have mentioned that you store a lot of your working context across sessions in some md file - what are you actually storing? What data do you actuall
43.
by austinbaggio 9mo ago
You store your project context in an ignored tmp folder? Share more plz - what does it look like? What do you store?
44.
by austinbaggio 9mo ago
It sounds like you've used quite a few. What programs are you expecting? Assuming you're talking about doing some inference on the data? Or optimizing for some RAG or something?
45.
by austinbaggio 9mo ago
In our experience, a lot of it is feel and dev preference. After talking to quite a few developers, we've found the skill was the easiest to get started with, but we also have a CLI tool and an MCP server too. You can check out the doc
46.
by austinbaggio 9mo ago
Since you've already thought about this problem, I'd love to hear your feedback after giving this skill a try. It should speed up at least your basic need of having to trigger the LLM to store the memory. One of our colleagues has
47.
by austinbaggio 9mo ago
Which of the 1000 is your favorite? There does seem to be a shallow race to optimizing xyz benchmark for some narrow sliver of the context problem, but you're right, context problem space is big, so I don't think we'll hurry
48.
by austinbaggio 9mo ago
I assume this is with work? And also assume you do send data, you just need some service agreement or something like with AWS or Microsoft for GH?
49.
by austinbaggio 9mo ago
Do you ever switch tools? I don't love the idea of my context being hostage of whatever LLM I choose first.
50.
by austinbaggio 9mo ago
It is overwhelming. We have support for Cursor mcp as well, but you lose a lot of the auto-magic stuff you get with the Claude Code plugin. Unfortunately, skills are pretty sticky to the Claude Code stack. It is kind of the vim of AI coding
51.
by austinbaggio 9mo ago
Good flag, we're still pretty early, I think the strict requirement for compensation disclosures is post 15 employees in CA? Did I get this wrong?
52.
by austinbaggio 9mo ago
You can use it with summaries for sure, but summaries often miss edge cases and long sessions drift. This makes it easier to jump between tasks, come back days later, and reorient without missing something that the summarization or compacti
53.
by austinbaggio 9mo ago
It feels like Claude is taking more of the Android approach of a less opinionated, but more open stack, so people are bending it to the shape they want to match their workflow. I think of the amnesia problem as pretty agent-agnostic, though
54.
by austinbaggio 9mo ago
Do you find you want to share the .md with the teams you work with? Or is it more for your solo coding?
55.
by austinbaggio 9mo ago
Honestly, very reasonable ask, you're not the first person to ask for a self-hosted version. We have a privacy policy we've drafted that is up-to-date with the current version of the product https://www.ensue-network.ai
56.
by austinbaggio 9mo ago
Appreciate it - yeah, you're right, models don't work well when you just give it a giant dump of memory. We store memories in a small DB - think key/value pair with embeddings Every time you ask Claude something, the skill: 1
57.
by austinbaggio 9mo ago
How are you storing the agent context now?
58.
by austinbaggio 9mo ago
Are you working with any projects to implement this yet?
59.
by austinbaggio 9mo ago
How are you finding it as context grows? I've been thinking a lot about the storage part of this - getting context to agents, temporal and semantic search, going as complex as graph implementations. . . overkill?
60.
Show HN: Stop Claude Code from forgetting everything(github.com)
202 points by austinbaggio 9mo ago|225 comments

More ›