resiros

7 ms·

31.
by resiros 11mo ago
Looks more of an aquisition for me. They are integrating the 50000 models into worker AI.
32.
by resiros 11mo ago
Author here. I wrote this blog post about our journey with COSS and the different models we experimented with. If you're running (or have run) a commercial open-source company, we'd genuinely love to hear your experience. The advi
33.
Commercial Open Source Is Hard: Our Journey(agenta.ai)
10 points by resiros 11mo ago|3 comments
34.
by resiros 11mo ago
TL;DR >I'll preface this by saying that neither of us has a lot of experience writing Python async code > I'm actually really interested in spending proper time in becoming more knowledgeable with Python async, but in our co
35.
by resiros 1y ago
There is a major mistake in the article. The author argues that openinference is not otel compatible. That is false. >OpenInference was created specifically for AI applications. It has rich span types like LLM, tool, chain, embedding, ag
36.
The AI Engineer's Guide to LLM Observability with OpenTelemetry(agenta.ai)
1 points by resiros 1y ago|0 comments
37.
Petition for a standardized pan-European 'EU–INC' corporate structure(eu-inc.org)
8 points by resiros 1y ago|5 comments
38.
by resiros 1y ago
Great work. I was expecting to use the informal transliteration keyboard, using 3 for ع or using 2 for ء or 7 for ح
39.
The AI Engineer's Guide to LLM Observability with OpenTelemetry(agenta.ai)
4 points by resiros 1y ago|0 comments
40.
by resiros 1y ago
Having been from the other side of the table. You did not flunk anything again. A job process is not an exam where if you do well you succeed. Your "performance" plays a small role in whether you are accepted (maybe less than 30%)
41.
by resiros 1y ago
Here is the report: https://mlq.ai/media/quarterly_decks/v0.1_State_of_AI_in_Bus... The story there is very different than what's in the article. Some infos: - 50% of the budgets (the one that fails) went to
42.
Building in Public: Why We're Publishing Our Roadmap(agenta.ai)
1 points by resiros 1y ago|0 comments
43.
by resiros 1y ago
For other open-source alternatives, check out https://github.com/agenta-ai/agenta (I am a maintainer), we're more focused on evals and collaboration between subject matter experts and engineers (friendlier UI, pow
44.
by resiros 1y ago
I look forward for the day someone trains a model that can do good writing, without emdashes, it's not but and all of the AI slop.
45.
by resiros 1y ago
> The endgame of that is an ecosystem of open source that ultimately serves absolutely nobody except for maybe startup PR needs. We self-host n8n, which by definition not open-source, and love it. It serves us. Not just n8n PR. That is t
46.
by resiros 1y ago
> For the flagship product of a company, especially a VC-funded company, not starting open source is the ethical thing to do. I can't even.. Seriously, I don't understand where your argument is coming from. Because, if you look
47.
by resiros 1y ago
I think that's perfectly fair. The community is quick to put whoever builds commercial OSS software on a cross the moment they change their license to ensure they still have a competitive advantage. Instead, we should encourage commerc
48.
Building Jetflow: a framework for performant data pipelines at Cloudflare(blog.cloudflare.com)
6 points by resiros 1y ago|0 comments
49.
by resiros 1y ago
Are you planning to open-source the benchmark environment and data (even anonymized) to allow people to compete on it. It looks like there are many ways to improve the accuracy of the agent by working on its logic (different tools, multi-ag
50.
by resiros 1y ago
That's a bummer, and I say this as a founder of a competing platform (Agenta). I think Humanloop had a lot of taste and were building one of the top products in the space. They were maybe the first company in the LLMOps space and set s
51.
by resiros 1y ago
Would love to understand it better too. It looks like the use cases are similar but the tech is different. NetBird is an alternative to Tailscale that uses Wireguard under the hood while these seem to use Traefik under the hood. I am person
52.
by resiros 1y ago
That's a big topic :D First thing about building AI or LLM powered apps is that 90% of the time you don't need to train models from scratch. LLMs are models that have the world's knowledge, so you can "train them" b
53.
by resiros 1y ago
Same observations we had talking to many of our users. The trick to build reliable systems, is minimize the complexity to the max, not the other way around. I think the theoretical value of multi-agents is collaboration with external agents
54.
by resiros 1y ago
1. Why is the author calling this agencies. He is talking about multi-agent systems, a space with research spanning decades (check https://en.wikipedia.org/wiki/Multi-agent_system ). Renaming this to agencies is weird.
55.
by resiros 1y ago
I think an approach like AlphaEvolve is very likely to work well for this space. You've got all the elements for a successful optimization algorithm: 1) A fast and good enough sampling function + 2) a fairly good energy function. For 1
56.
by resiros 1y ago
Skimmed quickly the paper. This does not look like RL. It's a genetic algorithm. In a previous life I was working on compbio (protein structure prediction), we built 100s of such heuristic based algorithm (monte carlo simulated anneali
57.
by resiros 1y ago
Most of the work when trying to build a reliable agent is 1) determining which tools do you provide to the llm 2) how to describe the tools, their interface. MCP gives you pre-built tools that you can't change neither the interface nor
58.
Ask HN: Is anyone building agents with MCP (or it's just for Claude integration)
1 points by resiros 1y ago|1 comments
59.
by resiros 1y ago
This looks very useful to get llms to generate slides. It would be nice if it had an llm.txt I can add as a context directly. Maybe an MCP server too
60.
by resiros 2y ago
I use personally for my home network. Very easy to use and quite mature. I'd highly recommend.

More ›