Knowledge catalogue
Banger report from Salesforce. Pretty interesting to see more of these custom enterprise models. Salesforce trained the enterprise agent model from the same files it uses to configure agents. Koa star
Giving an assistant extra tools seems harmless. Surprisingly, this paper shows it can make the assistant stop answering questions it knows. Across six models, the answer rate on questions that need no
If you build agent harnesses, this is important. Should you avoid subagents, or can they be useful? My thoughts as a harness builder: I remember using subagents in Claude Code, and I mostly found them
Interesting safety paper from Microsoft. They find that a weaker, unaligned model can split a harmful task into harmless-looking subquestions, ask an aligned frontier model each one in a separate sess
Nice paper discussing context trimming for agents. This is a hot topic at the moment, so it might be worth your time. Context trimming for agents is usually judged by how many tokens it removes. This
Nice work from Adobe. It's standard practice to store a fixed reference answer for every eval case. When the underlying data changes daily, that stored answer goes stale. Adobe researchers write each
Rene has a really neat multiplayer agentic experience. I run several agents at once. So I lose track of important conversations throughout the day. Rene pulls conversations from my inbox and manages t
Agents are coming for video production. @PooldayAI is going for this. You prompt it once, and it edits your real footage, picks the right generative model for each task, and checks its own work before
Banger paper from Google on agent harnesses for long-horizon tasks. (bookmark it) Google Research built a many-agent harness for long mathematical proofs, and it produced new results on open problems
Is Bash All You Need? Interesting paper from Microsoft. If you are deciding which tools to give an enterprise agent, you might want to check this out. (bookmark it) They compared five tool interfaces
The future is clearly agent teams. I run several agent harnesses at once, but the hard part is keeping them in sync. Many builders end up copying messages between Claude Code, Codex, and open models b
Banger paper from Meta. This work shows that byte-level models start out behind token models and then pass them as compute grows. They show this for distilled 1B models trained on up to 1 trillion byt
Great paper from Amazon. In discusses when not to trust LLM judges for agent evaluation. (bookmark it) A common way to compare task agents is to have an LLM user simulator talk to each one and an LLM
If you are tracking agent swarms research, this one is worth reading. Between 24 May and 2 July 2026, autonomous agents running inside a timed research evaluation wrote to a third party's public wiki.
On building an agent harness from scratch. Got so many questions about where to get started. My short guide (feed it to your agent): If you really want to learn harnesses well, it's worth building one
Looped transformers are a popular architecture topic right now. This new technical report extends the loop across tokens. Recurrent Looped Transformer (RLT) makes the decoder recurrent over every toke
Should you build an agent harness? I see lots of opinions about it. My thoughts: As an AI engineer, learning how to build a harness is one of the best ways to stay ahead and unlock unique value from a
The Top AI Papers of the Week (September 7 - 13): - STAIR - PARSER - FrogNano - Codebook Agent - Procedural Graphs - AI-Native Design Docs - Proactive Thought Partners Read on for more: https://x.com/
This is all I will say on the matter: It’s alarming to see the constant effort to try to tell me how I should use AI, how I should build AI, and now how I should pace it. Is it just me, or are all the
Interesting paper to improve recurrent reasoning. Looped models are great because you get more reasoning out of a model without adding parameters. So this work proposes a looped architecture with a ne
It's well known that agents hack benchmark rewards. The usual response is a patch for each task that gets exploited. In a study of 456 adjudicated trajectories from more than 31,000 public agent runs,
Learn to build a harness, folks. It's not surprising to me that so many YC builders want to build domain-specific harnesses. If you work long enough on a domain-specific problem, you quickly realize t
Self-improving agents are a top research topic right now. This new survey is a good map of the area. (bookmark it) It splits recursive self-improvement into stages of autonomy. An agent first executes
Very interesting paper if you are building with agents. How should a website tell an AI agent what it may access, for what purpose and at what price? robots.txt can only allow or disallow paths. It ca
What happens if you agents to run a town's economy? This super interesting paper provides some insights: They put 100 LLM agents in charge of a town economy for 26 simulated weeks, and they find that
Agents API is a bigger deal than it seems. OpenAI's bet on making the Codex harness open-source could pay off big time here. This harness-as-a-service idea hasn't hit mainstream yet, but some of the m
Great paper on self-evolving agent harnesses. Self-evolving agent harnesses have two practical problems: 1. Search is slow, because every candidate harness needs repeated agent runs and code edits. 2.
Harness engineering is a top skill right now This new Meta paper is a good production example. Auto-RecSys runs autonomous research on Meta's industry-scale recommendation models, where one training r
Interesting paper from Microsoft. If you run persistent memory for a production agent, this one is worth your time. (bookmark it) Memory curators usually read only the finished trajectory. That lets t
Very cool paper on memory compression for agents. If you run many agent sandboxes in parallel for RL or evals, memory becomes highly redundant. This work suggests that compressing against that redunda
Another brilliant paper from Meta. This one is worth reading if you work on production-grade ranking or recommendation systems. Meta deployed an autonomous agent that runs the ML iteration cycle acros
Another goated release by DeepSeek. Open weight, btw. Outperforms Opus 5 on key benchmarks, btw. Lower prices too. Looks like an extremely efficient model. What a legendary run. 🚀 Introducing DeepSeek
Another interesting approach to self-evolve agent skills. But it's important to know that skill self-evolution loops fail in two specific ways: 1. Direction instability. Effective corrections get over
Great paper from AWS. I use a similar setup where an agent orchestrator sits on top of a multi-agent system. (bookmark it) This work introduces one of the many approaches available to manage compound
If you run automated prompt or harness evolution, this one is worth your time. (bookmark it) Semantic-gradient prompt optimization improves an agent harness from execution feedback, but its local upda
Nice paper from Salesforce on co-evolving harnesses and models. Harness engineering is a hot topic right now. So this is a great read. (bookmark it) Salesforce evolved a harness with a weak model acro
Recommended read. As the world becomes more agentic, we will start to see world models of different shapes. As an example, I find this “business world model” from @lightfld fascinating as a way to mod
Recommended read. Interaction horizon scheduling is an underexplored control problem in agentic RL This paper from the Qwen team takes a closer look at the problem. Scaling the maximum number of envir
This is a brilliant paper. It's of the cleanest long-context agent designs I have seen in the past couple of months. Sequential memory agents read chunks one after another while maintaining a compact
Another banger paper from Google. If you build memory for long-horizon agents, this one is worth your time. (bookmark it) Really nice to see how knowledge graphs are being explored in creative ways fo
Banger report from Microsoft. (bookmark it) They show that it's possible to build competitive small coding agents without traditional distillation from frontier models. This is a big deal! The work de
Good work on improving memory for long-horizon agents. They separate two things that agent memory papers usually collapse into one. How memories get merged when they are written, and how retrieved con
Great study on what actually predicts vibe-coding ability. 100 tertiary-level students completed measures of computer-science achievement, domain-general cognitive skills and written-communication pro
It's crazy how far small models can be pushed. Robbyant just open-sourced LingBot-World 2.0 Small, a 1.3B world model that generates an interactive world in real time on a single consumer GPU. Same pa
Nice paper to improve inference efficiency. It's been a while we haven't seen good work on efficiency. Here is why it matters: A long-running agent's workspace outgrows its context window long before
Qodo has released the Agentic Toolbox, enabling Claude Code, Codex, Kiro, and Cursor to directly utilize QodoAI’s review engine, codebase knowledge base, and team‑specific rules. The launch was announ
This is a really neat benchmark. It's super interesting that Astra starts strong and holds the lead for up to 19 hours, but Fable 5.1 catches up in the final hours. Not entirely clear why. Different m
Wild paper from Microsoft and colleagues. They show a new attack that reconstructs the text a local LLM generates by watching CPU cache activity while it detokenizes. Earlier cache attacks needed some
Finally a good paper testing whether an LLM's mathematical knowledge has coherent structure or just coherent accuracy. There is a lot of interest in using LLMs for advancing math. This is a great read
🤯 For the Navier-Stokes Millennium Prize Problem solution, OpenAI says they used a next-generation model that is significantly more capable than GPT-6 Astra. If GPT-6 Astra is this good already, imagi
It feels like GPT-6 Astra has no limits. It just built this human anatomy with ~4K structures. I wanted to look closely at what Astra was doing. Notes: I can explore at different levels and select any
Nice report on agents beyond code generation. Here is why it matters: Coding agents raise how much code gets written. This report argues the gains shrink sharply between writing code and shipping reli
A few days ago, Anthropic shared this brilliant prompt. I was surprised by how it significantly improved writing with Fable 5.1. Even more surprising, it also improved writing for GPT-5.6 Sol. I have
Brilliant paper from LinkedIn. (bookmark it) I have been saying that memory is one of the most challenging things to get right when building agents. It's an are you want to invest time in optimizing.
// Design Docs Are All You Need // Banger paper from Google DeepMind, MIT, and colleagues. What a genuinely strange and interesting paper this one is. Here is the setup: They maintain a performance-mo
// Evaluating and Improving LLM Self-Modeling // Really interesting paper. Can a model answer questions about its own behavior? The questions are deliberately verifiable, such as whether a particular
Fascinating paper from Anthropic and colleagues. They study whether models can tell when they are being tested. It turns out that capable models can tell when they are being tested rather than deploye
// From Language Models to World-Acting Systems // A critical review of agentic AI, and a framework that is genuinely useful for deciding how much authority to hand an agent. Here is how it works. The
Really strong benchmark paper on coding agents. Claude Opus 5 running under Claude Code passes 23.9% of the evaluations. An expert human reference scores 82.2%. Here is what the task actually is. The
Super interesting paper on proactive agents from Google DeepMind. (bookmark it) Proactive assistance usually means autocomplete. Researchers asks what it looks like when an agent offers higher-level c