https://sparsenotes.com/categories/weekly-videos/ Recent content in weekly-videos on Sparse Notes https://sparsenotes.com/images/og-default.png https://sparsenotes.com/images/og-default.png Hugo -- gohugo.io Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/stroustrup-creator-of-cpp/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/stroustrup-creator-of-cpp/ Notes from a two-hour interview with the creator of C++ — what Bell Labs was actually like, why he insists abstraction can have negative overhead, the politics of memory-safety profiles, and the mistakes he wishes he'd fought harder against. https://sparsenotes.com/posts/2026/05/co-scientist-deepmind/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/co-scientist-deepmind/ DeepMind's six-minute overview of Co-Scientist — a multi-agent system that reads across the scientific literature, generates and ranks hypotheses, and gives working scientists an on-demand team of expert collaborators. https://sparsenotes.com/posts/2026/05/jane-street-ai-datacenter/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/jane-street-ai-datacenter/ Notes on Dwarkesh Patel's tour of Jane Street's Texas training cluster with Ron Minsky and Daniel Pontecorvo — 140 kW liquid-cooled GB300 racks, retrofit chillers, leak ropes, bus-bar power balancing, 8,000 km of fiber, and why opportunity cost dominates. https://sparsenotes.com/posts/2026/05/karpathy-lets-build-gpt/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/karpathy-lets-build-gpt/ Notes on Andrej Karpathy's 2023 classic 'Let's build GPT: from scratch, in code, spelled out.' — tokenization, the bigram baseline, the math of self-attention, multi-head, residual + layernorm, and how the same ~200 lines scale to a real GPT. https://sparsenotes.com/posts/2026/05/reliable-software-age-of-agents/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/reliable-software-age-of-agents/ Notes on Ron Minsky's talk on how Jane Street's long-standing bets — strong type systems, expect tests, library-level deterministic simulation, and serious code review — get more important, not less, in an agentic world, plus why the firm is suddenly interested in formal methods. https://sparsenotes.com/posts/2026/05/pi-to-pi-agent-orchestration/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/pi-to-pi-agent-orchestration/ Notes on IndyDevDan's walkthrough of peer-to-peer, bidirectional agent-to-agent communication built on the Pi coding agent harness — what changes when you flatten the orchestrator/worker hierarchy, when to use it, and the four-tool primitive that makes it work. https://sparsenotes.com/posts/2026/05/agent-development-lifecycle/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/agent-development-lifecycle/ Notes on Harrison Chase and Ankush Gola's Interrupt 2026 keynote — why agent development needs its own lifecycle, and the LangChain stack built around it (Deep Agents 0.6, LangSmith Sandboxes, Context Hub, SmithDB, LLM Gateway, LangSmith Engine). https://sparsenotes.com/posts/2026/05/transformer-vs-post-transformer/ Sat, 23 May 2026 00:00:00 +0000 https://sparsenotes.com/posts/2026/05/transformer-vs-post-transformer/ Notes from the Pathway-hosted 'Transformer vs Post-Transformer' panel — Łukasz Kaiser defending attention, with Adrian Kosowski (BDH/Pathway), Matthias Lechner (Liquid AI) and Llion Jones (Sakana AI) arguing for what comes next. Scaling laws, latent reasoning, hardware lock-in, benchmarks, and whether transformers themselves will discover their successor.