<aside> 💡
Pages serves as my latent space (hidden layer) for ideas and notes that aren’t yet mature enough (or are too brief) to warrant a full blog post.
PS: name is inspired from https://arxiv.org/abs/2412.06769
</aside>
Anatomy of vLLM | Aleksa Gordić
Inside NVIDIA GPUs (high-perf matmul) | Aleksa Gordić
🔗 how to read research papers | mason (stanford)
<aside> 💡
<github repo>
</aside>
May 1, 2025 → May 31, 2025
> Structured Decoding in language models
> Parallel Processing in Transformers
> Devstral: Design and implications
> Agents: Reflection vs ReAct
April 1, 2025 → April 30, 2025