GabrielBianconi

5 ms·

1. 1.▲
   by GabrielBianconi 3mo ago

   Thanks, appreciate the follow-up. It's certainly still to be determined if OSS AI infra will pan out, but I hope it does!
2. 2.▲
   by GabrielBianconi 3mo ago

   Our investors aren't looking for safe , they're looking for a small chance in funding the next Databricks or similar. Most times it doesn't work out unfortunately, but that's part of the game. (Also, we raised the capit
3. 3.▲
   by GabrielBianconi 3mo ago

   Thanks, that's exactly what happened. The title is misleading unfortunately but that's how social media goes...
4. 4.▲
   by GabrielBianconi 3mo ago

   Our team was much smaller. We didn't spend all the capital.
5. 5.▲
   by GabrielBianconi 3mo ago

   It's pretty common. If a startup winds down before it runs out of money, it typically returns whatever is left to the investors. We didn't have any debt.
6. 6.▲
   by GabrielBianconi 3mo ago

   Thanks!
7. 7.▲
   by GabrielBianconi 3mo ago

   See my sibling comment
8. 8.▲
   by GabrielBianconi 3mo ago

   There are many factors at play here but if I had to pick one... an open-source company has to find product market fit twice: first for the OSS project and again for a commercial product. The AI market moves very quickly so it's easy to
9. 9.▲
   by GabrielBianconi 3mo ago

   We raised most of the capital before we had any traction. We raised on a rolling basis and had millions in the bank before we had even published the open-source repository. Ultimately we raised based on the team's background + vision.
10. 10.▲
    by GabrielBianconi 3mo ago

    Yeah exactly. We didn't spend the majority of it.
11. 11.▲
    by GabrielBianconi 3mo ago

    Mostly salaries to support a small team. We are returning the remaining capital to investors.
12. 12.▲
    by GabrielBianconi 3mo ago

    Turns out I was wrong :)
13. 13.▲
    by GabrielBianconi 3mo ago

    Seed was in '24 actually but we only announced in '25.
14. 14.▲
    by GabrielBianconi 3mo ago

    We raised in 2024 and only burned through ~$3m of it, mostly on salaries to support a small team.
15. 15.▲
    by GabrielBianconi 3mo ago

    This was coincidental. Someone reported the issue last week, we fixed it, and published the advisory.
16. 16.▲
    by GabrielBianconi 3mo ago

    I'm the co-founder and CEO of TensorZero. We started the company two and a half years ago, and raised $7.3m in 2024 (announced only almost a year later). We've spent less than half of this amount. Earlier this week we came to the
17. 17.▲
    by GabrielBianconi 4mo ago

    Any function that can score (i.e. "evaluate") your LLM system (e.g. your agent). For example: - You write a heuristic (regex, code, etc.) that assigns a score to an output - You make another LLM score the output from your system (
18. 18.▲
    Even (very) noisy LLM evaluators are useful for improving AI agents(tensorzero.com)

    35 points by GabrielBianconi 4mo ago|10 comments
19. 19.▲
    Designing for Agents(twitter.com)

    1 points by GabrielBianconi 4mo ago|0 comments
20. 20.▲
    by GabrielBianconi 5mo ago

    It's getting more and more challenging to keep track!
21. 21.▲
    by GabrielBianconi 6mo ago

    TensorZero works with the OpenAI SDK out of the box: ``` from openai import OpenAI # Point the client to the TensorZero Gateway client = OpenAI(base_url="http://localhost:3000/openai/v1", api_key="not-used
22. 22.▲
    We're building an automated AI engineer, and it works(tensorzero.com)

    3 points by GabrielBianconi 6mo ago|0 comments
23. 23.▲
    Mitchell Hashimoto on Feature Design [video](twitter.com)

    3 points by GabrielBianconi 9mo ago|0 comments
24. 24.▲
    Bandits in Your LLM Gateway(tensorzero.com)

    3 points by GabrielBianconi 10mo ago|0 comments
25. 25.▲
    Claude Plays Catan [video](youtube.com)

    3 points by GabrielBianconi 1y ago|0 comments
26. 26.▲
    Is OpenAI's Reinforcement Fine-Tuning (RFT) Worth It?(tensorzero.com)

    4 points by GabrielBianconi 1y ago|0 comments
27. 27.▲
    How Kimi K2 achieves efficient RL parameter updates(moonshotai.github.io)

    3 points by GabrielBianconi 1y ago|0 comments
28. 28.▲
    Ask HN: How much would it cost to own and operate a personal gTLD?

    3 points by GabrielBianconi 1y ago|1 comments
29. 29.▲
    Deploying DeepSeek on 96 H100 GPUs(lmsys.org)

    285 points by GabrielBianconi 1y ago|80 comments
30. 30.▲
    Sporks of AGI: why the Real Thing is better than the Next Best Thing(sergeylevine.substack.com)

    3 points by GabrielBianconi 1y ago|0 comments