HN Explorer

kkm

↗ HN profile [ 77.8 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Aleph Alpha's latest Model playing Doom (tesseracted.com)
    1 points 1
  2. Aleph Alpha's Kolibri-1 playing Breakout – Making decisions under 25ms per move (twitter.com)
    1 points 0
  3. How to serve trillions of tokens for trillion-parameter coding agents (modal.com)
    1 points 0
  4. Unlocking parallel test-time scaling for long-horizon agents (blog.doubleword.ai)
    1 points 0
  5. Claude.md is good for taste and project context.It's a weak place for invariants (tesseracted-labs-blog.vercel.app)
    1 points 1
  6. Moving coding-agent guardrails from prompts to hooks (tesseracted-labs-blog.vercel.app)
    1 points 2
  7. Improving Throughput by Optimising KV Cache Efficiency for Agentic Workloads (j9s.io)
    1 points 1
  8. Guide to the Kimi DeltaNet Family of linear attention (blog.doubleword.ai)
    3 points 0
  9. Forensic Analysis of Container Snapshot Chains for Post-Event Reconstruction [pdf] (radostin.io)
    2 points 0
  10. The Agent swarm that designs itself (peterbhabra.com)
    1 points 0
  11. Don't Build a Router. Train the Small Model to Know When to Defer (distillabs.ai)
    2 points 1
  12. The gap between open weights LLMs and closed source LLMs (blog.doubleword.ai)
    306 points 30
  13. 1 points 0
  14. InfiniBand, RoCE, and All That (fergusfinn.com)
    5 points 0
  15. 2678x Faster Matrix Multiplication with a GPU (0mean1sigma.com)
    2 points 0
  16. UCCL-EP: DeepEP-style expert parallelism on any NIC, no GPU-initiated comms (fergusfinn.com)
    9 points 0
  17. Hacking Google with A.I. For $500k (brutecat.com)
    1 points 0
  18. How to setup a local coding agent on macOS (ikyle.me)
    507 points 57
  19. Anatomy of a high-performance EP kernel (fergusfinn.com)
    16 points 1
  20. No Token Left Behind: Demystifying Token-in-Token-Out in Miles (lmsys.org)
    2 points 0
  21. MoE expert co-activations: Reordering inputs yields easy throughput gains (blog.doubleword.ai)
    2 points 0
  22. The Economics of Speculative Decoding (fergusfinn.com)
    30 points 4
  23. Speculative KV coding: losslessly compressing KV cache by up to ~4× (fergusfinn.com)
    155 points 12
  24. 70x faster cold(ish) starts for SGLang (fergusfinn.com)
    1 points 0
  25. Bringing Up DeepSeek-V4-Flash on AMD MI300X (fergusfinn.com)
    120 points 11