Aydyn Tairov (@tairov) header
Aydyn Tairov (@tairov) profile picture

Production Engineer / ex-Meta

London
Joined March 2009
Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)Photo by Aydyn Tairov (@tairov)
You don't see many tweets starting with "We have proven a theorem..." these days. Congrats @machinestein
We have proven a theorem in complexity theory establishing a deterministic space bound strictly lower than the 50-year-old Savitch bound. Beyond pure theory, we used these proof constructions to build an inference engine that speeds up LLM generation on local models, especially on Apple Silicon. We show that on any graph, one can validate whether two points are connected in the few-path regime without a prior promise. While a quantum computer can validate this in logarithmic space, we prove that a classical computer can solve it in polynomial time and sub-Savitch deterministic space (O(log^2 M / log log M)).
1
1
138
Aydyn Tairov retweeted
Introducing Locals: an Apple Silicon engine that runs local LLMs up to 3x faster. It is built with a single mission: to free you from cloud providers and run frontier-class models locally without needing massive compute resources. It’s simple. Run our engine, and your coding harnesses and autonomous agents can immediately use it as a drop-in replacement for Ollama, MLX-LM, or llama.cpp. Under the hood is a theoretical breakthrough in complexity theory, enabling substantially faster algorithms for DeltaNet architectures, grammar-constrained decoding, tool calling, and structured outputs. Open source and available now on GitHub.
55
74
15
1,127
104,767
Aydyn Tairov retweeted
We present Tropical Reinforcement Learning. Standard RL for LLMs cannot learn compositions effectively, for a fundamental reason: its algebra is not optimal. It sums the probabilities of every successful answer, measuring how often a model succeeds, not which solution succeeds.
22
31
3
405
43,127
Migrated llama2.🔥 to Mojo 1.0 The port was easy using Claude, but the multithreading was broken hence the performance was terrible So I handed the same repo to Astra and Fable 5.1 and let them fix it... And here is what I got
1
13
674
I built pasteclaw .com - a pastebin for clawdbot / openclaw Install skill, or drop an AGENTS .md - ask your AI to make some HTML, get an instant live preview link Works with anything with CLI access including humans The agents needed their own pastebin. Now they have one.
1
215
I really hope Anthropic have finally hired a team of oncall KEs
Anthropic HQ must be in full freak out mode right now
4
464
Aydyn Tairov retweeted
While we are going back to the era of research… Introducing 𝗗𝗲𝗲𝗽 𝗜𝗺𝗽𝗿𝗼𝘃𝗲𝗺𝗲𝗻𝘁 𝗦𝘂𝗽𝗲𝗿𝘃𝗶𝘀𝗶𝗼𝗻 (𝗗𝗜𝗦) – a new learning method for recursive reasoning. DIS builds on the elegant Tiny Recursive Model (TRM)(@jm_alexia) but makes recursion radically simpler: - 𝟏𝟖× 𝗳𝗲𝘄𝗲𝗿 𝗳𝗼𝗿𝘄𝗮𝗿𝗱 𝗽𝗮𝘀𝘀𝗲𝘀 - 𝗡𝗼 𝗵𝗮𝗹𝘁𝗶𝗻𝗴 𝗺𝗲𝗰𝗵𝗮𝗻𝗶𝘀𝗺 - And a tiny 0.8M-parameter model reaching 24% accuracy on ARC-AGI-1 (@arcprize) Paper: arxiv.org/pdf/2511.16886 Code: github.com/machinestein/Deep…
1
6
22
5,489
Aydyn Tairov retweeted
Glad Anthropic bought the fast Js runtime.
Anthropic is acquiring @bunjavascript to further accelerate Claude Code’s growth. We're delighted that Bun—which has dramatically improved the JavaScript and TypeScript developer experience—is joining us to make Claude Code even better. Read more: anthropic.com/news/anthropic…
244
449
116
6,989
2,585,095
(1/9) When a 10-Line SQL Nearly Stopped the Internet On November 18th, Cloudflare's network (and half of the Internet) went down for nearly 3 hours The post-mortem just dropped, and the story is surprisingly simple... * No, it wasn’t DNS...
1
123
(8/9) 4️⃣ How did this get to prod? * The post-mortem doesn’t explain how the permission change in the critical part of the infra passed QA gates, code review, staging without raising a single red flag * It's not clear why at this scale CF don't have a Canary deployments
1
65
(9/9) The post-mortem write-up is a great case study in why resilience engineering matters more than ever
47