Yannick Sun (@yannick_sun) profile picture

e/acc | Foodie × Hiker × Hacker

San francisco
Joined July 2024
Yannick Sun retweeted
sharing my personal library on evals 1/n i put together the highest quality blogs, podcasts, papers, and projects on evals. additions are welcome!
28
278
18
2,287
234,823
Yannick Sun retweeted
must read for everyone who wants to reduce the entropy of their agentic systems
7
119
6
1,165
52,168
Yannick Sun retweeted
Fireside chat at Sequoia Ascent 2026 from a ~week ago. Some highlights: The first theme I tried to push on is that LLMs are about a lot more than just speeding up what existed before (e.g. coding). Three examples of new horizons: 1. menugen: an app that can be fully engulfed by LLMs, with no classical code needed: input an image, output an image and an LLM can natively do the thing. 2. install .md skills instead of install .sh scripts. Why create a complex Software 1.0 bash script for e.g. installing a piece of software if you can write the installation out in words and say "just show this to your LLM". The LLM is an advanced interpreter of English and can intelligently target installation to your setup, debug everything inline, etc. 3. LLM knowledge bases as an example of something that was *impossible* with classical code because it's computation over unstructured data (knowledge) from arbitrary sources and in arbitrary formats, including simply text articles etc. I pushed on these because in every new paradigm change, the obvious things are always in the realm of speeding up or somehow improving what existed, but here we have examples of functionality that either suddenly perhaps shouldn't even exist (1,2), or was fundamentally not possible before (3). The second (ongoing) theme is trying to explain the pattern of jaggedness in LLMs. How it can be true that a single artifact will simultaneously 1) coherently refactor a 100,000-line code base *and* 2) tell you to walk to the car wash to wash your car. I previously wrote about the source of this as having to do with verifiability of a domain, here I expand on this as having to also do with economics because revenue/TAM dictates what the frontier labs choose to package into training data distributions during RL. You're either in the data distribution (on the rails of the RL circuits) and flying or you're off-roading in the jungle with a machete, in relative terms. Still not 100% satisfied with this, but it's an ongoing struggle to build an accurate model of LLM capabilities if you wish to practically take advantage of their power while avoiding their pitfalls, which brings me to... Last theme is the agent-native economy. The decomposition of products and services into sensors, actuators and logic (split up across all of 1.0/2.0/3.0 computing paradigms), how we can make information maximally legible to LLMs, some words on the quickly emerging agentic engineering and its skill set, related hiring practices, etc., possibly even hints/dreams of fully neural computing handling the vast majority of computation with some help from (classical) CPU coprocessors.
Replying to @karpathy
@karpathy and I are back! At @sequoia AI Ascent 2026. And a lot has changed. Last year, he coined “vibe coding”. This year, he’s never felt more behind as a programmer. The big shift: vibe coding raised the floor. Agentic engineering raises the ceiling. We talk about what it means to build seriously in the agent era. Not just moving faster. Building new things, with new tools, while preserving the parts that still require human taste, judgment, and understanding.
378
807
148
6,204
1,368,721
Yannick Sun retweeted
Imagine every pixel on your screen, streamed live directly from a model. No HTML, no layout engine, no code. Just exactly what you want to see. @eddiejiao_obj, @drewocarr and I built a prototype to see how this could actually work, and set out to make it real. We're calling it Flipbook. (1/5)
1,133
3,719
1,168
28,864
6,091,472
What if AI could invent enzymes that nature hasn’t seen? 👩‍🔬🧑‍🔬 Introducing 🪩 DISCO: Diffusion for Sequence-structure CO-design 14 rounds of directed evolution and over a year of wet lab work. That's what it took to engineer an enzyme for selective C(sp³)–H insertion, one of the most challenging transformations in organic chemistry. DISCO surpasses this with a single plate. No pre-specified catalytic residues, no template, no theozyme, no inverse folding, just joint diffusion over protein sequence and structure. 📝 Blog: disco-design.github.io/ 📄 Paper: arxiv.org/abs/2604.05181 💻 Code: github.com/DISCO-design/DISC…
32
244
40
1,031
283,326
Yannick Sun retweeted
There are no words.
7,556
83,275
7,567
633,493
39,210,208
Yannick Sun retweeted
Approaching the near side of the Moon. The Artemis II astronauts have surpassed the record for the distance from Earth at 1:56 ET (1756 UTC). This record was previously set during the Apollo 13 mission when the astronauts traveled 248,655 miles from Earth. The Moon continues to grow larger and larger in the windows of the Orion spacecraft as the Artemis II mission gears up to observe the far side. The astronauts are predicted to make their closest approach of the Moon around 7:02pm ET (2302 UTC).
662
8,674
685
41,296
2,952,334
Yannick Sun retweeted
Walking alone through a foreign city at night and realizing how far you’ve come has to be a top 3 peak moment of all time
924
22,202
5,556
108,092
19,312,546
Yannick Sun retweeted
SOMEONE MADE A DIGITAL WHIP TO MAKE CLAUDE WORK FASTER 💀
1,594
11,231
2,442
141,246
14,930,416
Yannick Sun retweeted
Based on everything explored in the source code, here's the full technical recipe behind Claude Code's memory architecture: [shared by claude code] Claude Code’s memory system is actually insanely well-designed. It isn't like “store everything” but constrained, structured and self-healing memory. The architecture is doing a few very non-obvious things: > Memory = index, not storage + MEMORY.md is always loaded, but it’s just pointers (~150 chars/line) + actual knowledge lives outside, fetched only when needed > 3-layer design (bandwidth aware) + index (always) + topic files (on-demand) + transcripts (never read, only grep’d) > Strict write discipline + write to file → then update index + never dump content into the index + prevents entropy / context pollution > Background “memory rewriting” (autoDream) + merges, dedupes, removes contradictions + converts vague → absolute + aggressively prunes + memory is continuously edited, not appended > Staleness is first-class + if memory ≠ reality → memory is wrong + code-derived facts are never stored + index is forcibly truncated > Isolation matters + consolidation runs in a forked subagent + limited tools → prevents corruption of main context > Retrieval is skeptical, not blind + memory is a hint, not truth + model must verify before using > What they don’t store is the real insight + no debugging logs, no code structure, no PR history + if it’s derivable, don’t persist it
151
680
96
6,269
837,267
Yannick Sun retweeted
New AI paper from us this week. When my student first showed me his initial findings, I really didn’t know what to make of them. I felt that this was an interesting but curious loophole phenomenon that would shortly be closed. I was very wrong. arxiv.org/abs/2603.21687
59
255
72
1,488
479,909
Yannick Sun retweeted
The AI Scientist: Towards Fully Automated AI Research, Now Published in Nature Nature: nature.com/articles/s41586-0… Blog: sakana.ai/ai-scientist-natur… When we first introduced The AI Scientist, we shared an ambitious vision of an agent powered by foundation models capable of executing the entire machine learning research lifecycle. From inventing ideas and writing code to executing experiments and drafting the manuscript, the system demonstrated that end-to-end automation of the scientific process is possible. Soon after, we shared a historic update: the improved AI Scientist-v2 produced the first fully AI-generated paper to pass a rigorous human peer-review process. Today, we are happy to announce that “The AI Scientist: Towards Fully Automated AI Research,” our paper describing all of this work, along with fresh new insights, has been published in @Nature! This Nature publication consolidates these milestones and details the underlying foundation model orchestration. It also introduces our Automated Reviewer, which matches human review judgments and actually exceeds standard inter-human agreement. Crucially, by using this reviewer to grade papers generated by different foundation models, we discovered a clear scaling law of science. As the underlying foundation models improve, the quality of the generated scientific papers increases correspondingly. This implies that as compute costs decrease and model capabilities continue to exponentially increase, future versions of The AI Scientist will be substantially more capable. Building upon our previous open-source releases (github.com/SakanaAI/AI-Scien…), this open-access Nature publication comprehensively details our system's architecture, outlines several new scaling results, and discusses the promise and challenges of AI-generated science. This substantial milestone is the result of a close and fruitful collaboration between researchers at Sakana AI, the University of British Columbia (UBC) and the Vector Institute, and the University of Oxford. Congrats to the team! @_chris_lu_ @cong_ml @RobertTLange @_yutaroyamada @shengranhu @j_foerst @hardmaru @jeffclune
74
462
98
2,453
975,715
Yannick Sun retweeted
we just killed writer’s block. introducing nimbus, now you can write at the speed of thought. real time auto-research and compounding intelligence, no prompting required.
26
23
3
120
7,622
Yannick Sun retweeted
Introducing SkillsBench - also call for open-source contributors SkillsBench is the first benchmark that tests whether agent skills can improve performance, and how well agents use skills. If you are interested, check out the GitHub in the comment, Dm, and join our Discord
12
36
9
291
57,191
Hi Twitter 👋 We’re Evolved — organizers of the AI × Bio Hackathon (Nov 7–16). Today we’re announcing our judges: leaders from NVIDIA, OpenAI, FutureHouse, EvolutionaryScale, and more shaping the future of AI in biology.
2
11
5
27
20,116
At FutureHouse, we’ve noticed scientific agents are good at applying average intelligence across tasks. They always seem to make the obvious choices, which is good, but discovery sometimes requires more intuition and insight than average. We’ve made the first step today towards superhuman insight by training a reasoning model for a specific domain of science: designing drug-like molecules. We’re releasing a 24B open-weights reasoning model called 𝚎𝚝𝚑𝚎𝚛𝟶. 𝚎𝚝𝚑𝚎𝚛𝟶 has been trained with reinforcement learning to exceed frontier and human experts across a range of molecular design tasks. 𝚎𝚝𝚑𝚎𝚛𝟶 takes in natural language, reasons in English, and outputs a new molecule. 𝚎𝚝𝚑𝚎𝚛𝟶 is now a tool for our chemistry design agent, Phoenix, which can call upon it to design molecules. Training a reasoning model for a scientific domain like chemistry, rather than math or programming, required a number of small technical advances. For example, we developed an iterative method of split specialist models and aggregation of reasoning traces. Another example is we used LLMs to rewrite questions that were partially solved. A major finding from this work is that we can train with >10x efficiency per experimental measurement when using a reasoning model, rather than fine-tuning. We also found that reasoning models can learn new tasks, developed specifically for this paper and not in pretraining corpora. We even saw a task have 0% performance until 100 steps into RL, at which it randomly solved once. This, along with our change in modality from natural language to molecules, bodes well for applying reasoning models far from natural language. Reasoning models in science are the future. Scientific tasks are naturally verifiable rewards: the physical world is the ultimate arbiter of accuracy, rather than human contractors. The data efficiency gain and ability to exceed frontier models with relatively few parameters/compute mean that we should expect more scientific reasoning models soon. Congrats to team @SidN137, James, @Ryan__Rhys, Albert, @GWellawatte , @maykcaldas , @ludomitch , and @SGRodriques. Thanks to @VoltagePark @nvidia and @huggingface for supporting us, and huge thanks to @ericschmidt for funding @FutureHouseSF The model weights, reward model, and new benchmark are open source. You can also read more about scientific reasoning models in our exclusive with Nature.
13
79
17
405
80,953
Yannick Sun retweeted
Does RAG solve hallucination? Even w/ RAG, we found that >30% of LLMs' medical statements are not fully supported by (sometimes contradict) the cited refs @NatureComms nature.com/articles/s41467-0… We present #SourceCheckUp agent to verify faithfulness of LM to source info. Great job @kevinywu @ericwu93 w/ awesome collaborators👏
12
69
7
322
31,768