keremk

I suffer from the same problem of marketing for my own open source tool that I built. So now I am attempting to use AI to help me market it. But instead of creating a fully automated harness that generates content on its own and posts it to various places, and gets flagged as slop AI content, it helps me identify the places to post by finding relevant discussions in relevant forums, places. Then I draft my own message. In fact it found me this thread here, because I am also working on a tool (for AI films in this case, not product videos but still adjacent). You can ask Claude (or Codex) to create yourself a similar thing. Mine wakes up every morning and finds me places to post, tracks it outcomes etc. Gives me a Kanban board as well to keep track of progress.

Opus 5.5 and Astra are already very good (because of being RL trained at it) most of these. The way I see skills useful are: 1)If they are created to teach how to use a custom and new CLI or MCP 2) They describe gotchas, tips and tricks that are hard to find using simple search or potentially save some tokens for easier to find ones. The OpenAI and Anthropic teams already said to reduce the skills otherwise as they can also waste context space unnecessarily.

Occam's Razor says: "They heard this problem is solved or about to be solved amongst the rest of the other problems. They prioritized this and put substantial compute with their newest model and solved it." I know everyone loves juicy rumors, theories etc. but honestly that is the simplest and most plausible explanation given the state of AI improvement now. Obviously spending 15 million on a problem is not a slam dunk decision even for a company like OpenAI but if it has a significantly high chance of solving it and their competitor will be claiming they solved it, then it raises the stakes and they go after it. In fact this is the most rational and also curiosity-driven thing to do and totally what I would have expected from any frontier lab. Of course if one wants to prove their confirmation biases that they train on sessions or be able to identify individual users, the non-zero chance of that being also another explanation is attractive enough to wet their appetites.

For me, Opus 2.6 has been a huge regression. It now hangs for 10+ minutes on a task that it used to take a few minutes to complete, and all I see is some "Reading 3 files" etc. messages showing nothing else. The 2 issues where Opus 2.6 is absolutely not as good as 2.5 at all by miles and also showing just some obscure unnecessary messages makes me say "why?" Why did you decide to screw something up that was awesome and dumb it down and keep insisting that the "verbose" mode is the way to go?? Seriously who wants to see messages that are essentially the same? 2 patterns, 3 files read?? Seriously? Who is that mode for? Why is it even a default??

Actually this is how LLMs (with reasoning) work as well. There is the pre-training which is analogous to the human brain getting trained by as much information as possible. There is a "yet unknown" threshold of what is enough pre-training and then the models can start reasoning and use tools and the feedback from it to do something that resembles to human thinking and reasoning. So if we don't pre-train our brains with enough information, we will have a weak base model. Again this is of course more of an analogy as we yet don't know how our brains really work but more and more it is looking remarkably aligned with this hypothesis.

For these kinds of blogposts where the content is very good but not very approachable due to assumption of extensive prior knowledge, I find using an AI tool very useful, to explain and simplify. Just used the new browser Dia for this, and it worked really well for me. Or you can use your favorite model provider and copy and paste. This way the post stays concise, and yet you can use your AI tools to ask questions and clarify.