The week ahead: 20251201
[Blog] HOPE. A model as a set of nested sub-optimizers, each with its own context flow and hyperparams. Relevant paper.
Writing is an important way to organize one’s thoughts. If you can’t express something, you probably haven’t thought it through.
I write with two goals:
- to reflect on what I read and learn;
- to make useful information accessible to people who need it.
With these, I consider brevity a writer’s highest virtue.
Recent Posts
The week ahead: 20251124
[Tweet] Sergey’s interview.
The week ahead: 20251117
[GitHub] Karpathy’s EPUB reader for vibe reading. Revelant tweet.
The week ahead: 20251110
[PEP] Python finally officially supports lazy imports!
The week ahead: 20251103
[Article] ChatGPT for mental health.
The week ahead: 20251027
Q3 Earnings Week!
The week ahead: 20251020
[Paper] Synthetic customers for purchase intent prediction. Rather than generating a Likert score for intent (and there would be many relevant techniques ...
The week ahead: 20251013
[Paper] Embedding initialization shown to be related to Grokking (delayed generalization). On the XOR task, embeddings trained on a small model is shown t...
The week ahead: 20251006
[Blog] Thinking Machines Lab: LoRA is better to be applied to all weights, including MLP and MoE. On RL, LoRA performs on par with full training even with...
The week ahead: 20250929
[Announcement] ChatGPT Pulse. ChatGPT can now proactively push personalized content.
The week ahead: 20250922
[Blog][Tweet] Gemini (10/12) and GPT (12/12) compete at ICPC.
The week ahead: 20250915
[Blog] First glimpse into the work of Thinking Machines Lab. Inference systems can be made deterministic by using “batch-invariant” GPU kernels (GitHub im...
The week ahead: 20250908
In NYC for Gemini summit.
The week ahead: 20250901
[Blog] GPT-oss from the ground up.
The week ahead: 20250825
Traveling in Brazil this week to join the indigenous Xingu people for Quarup.
The week ahead: 20250818
[Reuters] Perplexity makes $34.5B bid on Google’s Chrome.
The week ahead: 20250811
[HuggingFace] OpenAI’s open-weight model. Technical report from OpenAI. Onboarding guide from HuggingFace.
The week ahead: 20250804
[Blog] Utkarsh’s insights on the mathematical impossibility of some expectations of AI agents in 2025. TLDR: error compounding, token economics, tool engi...
The week ahead: 20250728
[Report] Alphabet Q2 earnings. CapEx increases for the 2nd time in this year.
The week ahead: 20250721
[Blog] Perplexity to launch its own browser, with rumors of OpenAI’s upcoming release of a similar product, signaling challenges to Google Chrome’s contro...