HN Explorer

kcorbitt

↗ HN profile [ 75.7 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Codex, File My Taxes. Make No Mistakes (corbt.com)
    5 points 1
  2. A Pocket Guide to Surviving the Robot Apocalypse (corbt.com)
    2 points 0
  3. Show HN: RULER – Easily apply RL to any agent (openpipe.ai)

    Hey HN, Kyle here, one of the co-founders of OpenPipe. Reinforcement learning is one of the best techniques for making agents more reliable, and has been widely adopted by frontier labs. However, adoption in the outside…

    81 points 8
  4. Everything I know about reward hacking (openpipe.ai)
    3 points 0
  5. Show HN: ART – a new open-source RL framework for training agents (github.com)

    Hey HN, I wanted to share a new project we've been working on for the last couple of months called ART (https://github.com/OpenPipe/ART). ART is a new open-source framework for training agents using reinforcement…

    116 points 13
  6. ART·E: how we built an email research agent that beats o3 (openpipe.ai)
    3 points 3
  7. Using GRPO to Beat o1, o3-mini and R1 at “Temporal Clue” (openpipe.ai)
    199 points 59
  8. Analyzing OpenAI's Reinforcement Fine-Tuning: Less Data, Better Results (openpipe.ai)
    4 points 0
  9. Using reinforcement learning and $4.80 of GPU time to find the best HN post (openpipe.ai)
    212 points 97
  10. Show HN: Agent.exe, a cross-platform app to let 3.5 Sonnet control your machine (github.com)
    406 points 243
  11. DPO fine-tuning outperforms SFT (openpipe.ai)
    1 points 0
  12. OpenPipe Mixture of Agents: Outperform GPT-4 at 1/25th the Cost (openpipe.ai)
    13 points 2
  13. What we've learned in 3 days of Llama 3 (openpipe.ai)
    3 points 1
  14. Mixtral Curious? Comparing Mistral 7B and Mixtral for fine-tuning (openpipe.ai)
    1 points 0
  15. S-LoRA: Serving Thousands of Models from One GPU for Fun and Profit (openpipe.ai)
    1 points 0
  16. Is AI the next crypto? Insights from HN comments (openpipe.ai)
    237 points 381
  17. Fine-tune your own Llama 2 to replace GPT-3.5/4

    There has been a lot of interest on HN in fine-tuning open-source LLMs recently (eg. Anyscale's post at https://news.ycombinator.com/item?id=37090632). I've been playing around with fine-tuning models for a couple of…

    955 points 192
  18. Show HN: Automatically convert your GPT-3.5 prompt to Llama 2

    Hey HN! I'm working on OpenPipe, an open source prompt workshop. I wanted to share a feature we recently released: prompt translations. Prompt translations allow you to quickly convert a prompt between GPT 3.5, Llama 2,…

    13 points 2
  19. TaxyAI: Open-source browser automation with GPT-4 (github.com)
    355 points 105
  20. Tell HN: YC will help you find a co-founder

    Hey HN, I’m Kyle Corbitt, and I work on Startup School (https://www.startupschool.org/), YC’s free program to help people learn how to start a startup. Today we’re launching a new major feature: co-founder matching…

    433 points 136
  21. YC Startup School for future founders who aren't quite ready to start yet (blog.ycombinator.com)
    330 points 92
  22. YC's Startup School Relaunching as Continuous Program (blog.ycombinator.com)
    330 points 60
  23. Y Combinator is hiring an engineer to help us scale Startup School (jobs.lever.co)
    1 points 0 job
  24. As Uber and Tesla struggle with driverless cars, Waymo moves forward (arstechnica.com)
    76 points 49
  25. McKinsey: One-third of US workers could be jobless by 2030 due to automation (cnbc.com)
    5 points 1