# billion tokens (blogs) — RSS Amplifier Recent posts from the 2 feeds in the RSS Amplifier directory that cover billion tokens. Page: Feed: --- ## [Talking to AI is terrible](https://gusarich.com/blog/talking-to-ai-is-terrible) _2026-08-17 · Daniil Sedov — Blog_ ## [Retrospective: spending 500 billion Codex tokens](https://adamadam.blog/2026/06/10/retrospective-spending-500-billion-codex-tokens/) _2026-06-10 · Adam Zieliński · Adam's Perspective_ For the last month, I ve had unlimited access to OpenAI Codex. It s been such a weird experience! This post is about my experience, learnings, and reflections. What I ended up shipping Quite a lot. I m sharing this early in the post to put the rest of it in perspective. It took 500 billion tokens in \[ \] ## [General, automated WordPress-to-WordPress sync is unsolvable](https://adamadam.blog/2026/06/10/general-automated-wordpress-to-wordpress-sync-is-unsolvable/) _2026-06-09 · Adam Zieliński · Adam's Perspective_ Twenty-three years into WordPress, the general problem of site synchronization is still unsolved. We have great tools for the slices, but nothing for the whole. You can t reliably automate any of these tasks beyond basic sites: This post exhaustively covers what makes these problems so difficult, and what we can do instead. It is a \[ \] ## [Running 80 concurrent Codex sessions](https://adamadam.blog/2026/05/26/running-80-concurrent-codex-sessions/) _2026-05-26 · Adam Zieliński · Adam's Perspective_ With unlimited tokens, the constraint becomes compute power: CPU, memory, disk space. Dan Luu s and Dennis Snell s work inspired me to move from managing Codex sessions to assigning it a big task and a pool of compute resources to maximize. It burns tokens! I ve got through ~120 billion tokens since Thursday. Dan is at ~300 \[ \] ## [AI, generate 100 designs for WordPress Playground](https://adamadam.blog/2026/05/22/ai-generate-100-designs-for-wordpress-playground/) _2026-05-21 · Adam Zieliński · Adam's Perspective_ I wanted to revamp WordPress Playground UI for a while but I couldn t get any useful results from LLMs. Until today. I ve shared my struggle on Slack and Chris Huber inspired me to ask it not for 100 iterations on one design, but for 100 different designs. So I did. And I got 100 designs. \[ \] ## [My thoughts on building software using AI agents](https://adamadam.blog/2026/05/13/my-thoughts-on-building-software-using-ai-agents/) _2026-05-13 · Adam Zieliński · Adam's Perspective_ I use LLMs a lot lately. I even got an unlimited access to a popular commercial model recently, which increased my usage even more. It s all still so new to me. I keep poking at the limits and discovering what it means for software and three things come up for me on most days: Human \[ \] ## [Turn your handwriting into a font](https://adamadam.blog/2026/04/02/turn-your-handwriting-into-a-font/) _2026-04-02 · Adam Zieliński · Adam's Perspective_ Jamie Marsland published this great video on how to turn your handwriting into a font, and then how to use that font on your WordPress site. Check it out! ## [My notes from „Butterick’s Practical Typography”](https://adamadam.blog/2026/04/01/my-notes-from-buttericks-practical-typography/) _2026-04-01 · Adam Zieliński · Adam's Perspective_ I’ve just refreshed Butterick’s Practical Typography. The first time I read it was a couple of years ago and I’ve surely forgotten a bunch since. Here are a few thoughts I had during this read-through. Typography is like theater Butterick cites three principles from The Elements of Typographic Style: The three principles read exactly like \[ \] ## [SQL Isolation levels: Foundational database feature most developers don’t know about](https://adamadam.blog/2026/03/26/sql-isolation-levels-foundational-database-feature-most-developers-dont-know-about/) _2026-03-26 · Adam Zieliński · Adam's Perspective_ I looked at SQL queries through the lens of isolation levels ever since my brother showed me that concept about 16 years ago. I ve noticed many, many senior devs are unaware of this fundamental database feature. Here is a short primer on isolation levels that every backend developer should read. The core problem Suppose we \[ \] ## [My LLM Tier List](https://gusarich.com/blog/my-llm-tier-list) _2026-02-05 · Daniil Sedov — Blog_ ## [Recipe PDFs Ready Fast (Sponsored)](https://crawlproof.com/a/krOhXbB5MSfx) _2026-02-05 · **Sponsored**_ Turn online or personal recipes into clean, printable PDFs in seconds. ## [Things got too easy with AI](https://gusarich.com/blog/things-got-too-easy) _2026-01-27 · Daniil Sedov — Blog_ ## [Do less with AI. And breathe.](https://adamadam.blog/2026/01/21/do-less-with-ai-and-breathe/) _2026-01-20 · Adam Zieliński · Adam's Perspective_ Waiting for AI to finish is uncomfortable. I found I feel better and get more done when I stay with that discomfort without acting on it. When, instead of rushing to open my fourth concurrent agent or reply to another Slack ping, I just stay right there, present. Maybe I’ll close my eyes for a moment and focus \[ \] ## [I gave Codex its own Mac Mini](https://gusarich.com/blog/i-gave-codex-its-own-mac-mini) _2026-01-08 · Daniil Sedov — Blog_ ## [TON Vanity: 286,000x faster vanity addresses](https://gusarich.com/blog/ton-vanity) _2026-01-02 · Daniil Sedov — Blog_ Rebuilding vanity address generator for TON from scratch with AI, achieving up to 286,000x speedup through a series of domain-specific optimizations. ## [AI in 2026](https://gusarich.com/blog/ai-in-2026) _2025-12-31 · Daniil Sedov — Blog_ Predictions and thoughts on AI progress in 2026. ## [WordPress Playground: 2025 Year in Review](https://adamadam.blog/2025/12/04/wordpress-playground-2025-year-in-review/) _2025-12-04 · Adam Zieliński · Adam's Perspective_ This post is re-published post from https://make.wordpress.org/playground/2025/12/03/wordpress-playground-2025-year-in-review/ A lot happened for Playground this year! Let s review what changed and why it matters for your work: 99+% of WordPress Plugins Supported in Playground Nearly every plugin from the WordPress plugin directory works in Playground. In testing with the top 1,000 plugins, 99%… ## [What LLM to use today?](https://gusarich.com/blog/what-llm-to-use-today) _2025-11-28 · Daniil Sedov — Blog_ ## [There is nothing out-of-distribution](https://gusarich.com/blog/there-is-nothing-out-of-distribution) _2025-11-20 · Daniil Sedov — Blog_ ## [There is no singularity](https://gusarich.com/blog/there-is-no-singularity) _2025-10-12 · Daniil Sedov — Blog_ ## [Writing with AI](https://gusarich.com/blog/writing-with-ai) _2025-10-12 · Daniil Sedov — Blog_ ## [AI QA That Opens Fix PRs (Sponsored)](https://crawlproof.com/a/Yqw0csN6jvxc) _2025-10-12 · **Sponsored**_ Runs browser QA, files issues with repro evidence, and opens fix PRs. ## [My impression of GPT-5](https://gusarich.com/blog/my-impression-of-gpt-5) _2025-08-11 · Daniil Sedov — Blog_ ## [The complexity threshold of AI](https://gusarich.com/blog/the-complexity-threshold-of-ai) _2025-07-29 · Daniil Sedov — Blog_ ## [Billions of Tokens Later: Scaling LLM Fuzzing in Practice](https://gusarich.com/blog/billions-of-tokens-later) _2025-07-18 · Daniil Sedov — Blog_ Lessons learned from scaling documentation-driven black-box fuzzing pipelines to billions of tokens, practical deduplication strategies, discovered scaling laws, and initial explorations into white-box fuzzing for future expansion. ## [Multitasking in 2025](https://gusarich.com/blog/multitasking-in-2025) _2025-03-30 · Daniil Sedov — Blog_ ## [Documentation-Driven Compiler Fuzzing with Large Language Models](https://gusarich.com/blog/fuzzing-with-llms) _2025-03-26 · Daniil Sedov — Blog_ A fresh and simple black-box approach to fuzzing compilers using large language models to generate test cases from documentation and specification. ## [Measuring and Analyzing Entropy in Large Language Models](https://gusarich.com/blog/measuring-llm-entropy) _2025-03-11 · Daniil Sedov — Blog_ A detailed benchmarking study exploring entropy and randomness across 52 large language models using diverse prompting strategies, revealing notable biases and significant variability influenced by model architectures and prompt engineering.