ricky (@drf0k) header
ricky (@drf0k) profile picture

building the duolingo for staying active @stridan_app | building hackathons at hackerfund | @ucla | @nasa | backed by god

san diegooooo
Joined June 2018
Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)Photo by ricky (@drf0k)
ricky retweeted
Unpopular opinion time, but anybody that actually codes and works in a development industry, and actually read this article, would agree this doesn’t matter at all. Please stop punishing developers for using code assist tools… Punish them for generating music and art.
Big Walk used AI in 'the final stretch' of development, kicking off intense awards debate: aftermath.site/big-walk-ai-i…
Readers added context they thought people might want to know
House House used LLMs sparingly near the end of Big Walk's development, mainly to fix bugs in third-party code. They estimate well under 0.1% of the game's code was affected, and they didn't use generative AI elsewhere. bigwalk.game/faq/#ai pcgamer.com/games/puzzle/b…
200
944
51
12,166
333,539
Some thoughts on building in 2026:
Today, we're shipping real-time graphics. Using Sol 6.1 Ultrafast we can generate custom overlays, diagrams, and even interactive 3D scenes right on a user's screen, no chat. It's a new visual interface for AI. Here's me learning about LLMs + hardware. Demo:
37
43
16
1,062
65,689
ricky retweeted
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected. Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions. Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent. But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models? We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem. Read the full report: cohere.com/blog/concerning-b…
974
1,282
1,522
11,086
3,106,579
ricky retweeted
Today, we announce our $25M fundraise to build the future of creative work. For all of history, what you could create was dictated by the resources accessible to you. That's why we built @trymelius, the agents lab for creatives, to tear down that constraint and let agents leverage generative models to bring all your ideas to life. So that anyone can make anything. Huge thanks to @yuris at @generalcatalyst and @verorellana02 + @reidRMC at @CRV for leading the round, and to @gefkovicz + @benmtaft at @geniusventures, @amosk1120 + @adamvalkin at @VineVenturesLP, @jakepaul + @geoffwoo at @antifund, @rauchg, @bykahlil and many more amazing investors for taking a bet on us.
274
99
104
1,612
870,176
I've decided to make my Midjourney account public. 🖤 midjourney.com/@ciguleva?tab… Wander through the archive, borrow any style that speaks to you, and use the images freely. Midjourney has given me so much joy and taught me more than I ever expected. If my work can spark something in you or help you on your own creative path, that would truly make my day. I'm also opening up mjpro(dot)ai completely, for free. To my early subscribers: thank you! You believed in this from the very beginning, and you won't be charged again. If you ever have questions, have concernes or just want to say hi, my inbox is always open. Happy creating ✨
100
94
23
1,773
152,482
We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.
1,709
6,773
1,588
56,907
8,254,759
many people ask when to use a big model (like sol/fable) at low reasoning effort, vs a small model (like luna/sonnet) at high reasoning effort i deliberately forced myself to use all the permutations a lot over the last couple of weeks to build intuition, and i realized the difference is "wisdom" vs "diligence" bigger models are "wiser" they have seen a lot. they remember a lot. they have a lot of expertise across different domains. they have better intuition, can connects the dots, and come up with creative, inspired ideas reasoning effort makes a model more "diligent" it'll assess each option, think through consequences, and figure out edge cases etc more thoroughly if there are 100 paths ahead, diligence makes the model assess every single one without a miss, but it will not make the model realize maybe the best one is to take none of the 100 paths and instead dig a tunnel "wisdom" and "diligence" are orthogonal. and now i get why the models are launched the way they were, and not just a single fable level model with 12 different reasoning levels we're all misled by the way we've been plotting the models with benchmark scores, which are fundamentally flawed because they use a single dimension to measure the model's capability, making us think of model size and reasoning effort as being fungible with each other, while in fact "wisdom" and "diligence" needs to be measured separately i hope the evals eventually catch up and address this. until then, here's my recommendation for how to choose - - if the problem you are trying to solve is something you think requires a genius, use a bigger model - if the problem you are trying to solve is something you think requires a pen and lots of paper, use higher reasoning effort - if it requires a genius sitting down with a pen and lots of paper, tune up both
91
108
41
1,731
145,447
Love the tastefulness with which anthropic ships. Main reason I got excited about 'headless' for the web is that any given page could take whatever whimsical or unique shape you wanted. With AI, there's no excuse not to push the design frontier. This is @nextjs at its finest: anthropic.com/claude-opus-5-…
74
34
5
1,688
99,603
i'm hiring a senior jev engineer at least 2 years of experience with jev is a requirement dm
429
285
76
8,065
258,151
ricky retweeted
yo @aicrowdHQ @agihouse_org my engineer placed 1st in your post-training competition 8 months ago and still hasn't been paid his prize yet i believe y'all owe him 15 bands, pay him out
3
1
79
5,937
ricky retweeted
Hi Astra users. A reset and a quick update on quality issues that have been posted around. Working with some of you, we have found and fixed the following issues: - Some skills written for previous models were triggering too often or preventing the model from checking its work. - An opt-in context management experiment that could cause early stops or replies to older messages. We've disabled it. Our rough estimate is that 4-5k users were affected by this experiment. - We've also removed some badly configured engines that resulted in a measured quality degradation for a long tail of traffic flowing through them. We’ve also made some more minor improvements and things should feel significantly better across the board. More consistent follow-through, better tracking of your latest message, and better checks on the work as it’s going through the motions. The examples posted and all the users who worked directly with us were incredibly useful in helping fix things quickly. Always grateful for this incredible community. And of course, a reset is also landing by midnight today.
3,911
1,589
1,969
28,203
8,626,923
ricky retweeted
Because we are beyond happy to have Astra rolled out today ahead of schedule and you have been super patient with us (not really, but it’s ok!)… we will do the full banked reset today too for all Plus, Pro and Business users. Lands end of day. Happy Astra day and enjoy a phenomenal weekend. PS: If you create the account or upgrade before 8pm PT you will get it too. Still time!
2,177
1,043
683
25,222
2,520,825
ricky retweeted
Here's an example. A week ago I asked an agent (frontier model) to watch a restaurant and let me know when reservations open for Dec 20. Said okay, it will ping me. A few days later I checked (just to make sure it was working). "Still closed." Two days later, same asnwer. Then today it came back with an update. "Dec 20 won't open. The restaurant is closed that Sunday." That didn't sound right. So I checked the site myself. Reservations hadn't opened. Dec not even selectable on the calendar. I asked the agent how it checked Dec 20. It said it forced Dec 20 through the url, got a closed response, concluded the restaurant was closed that day. It cheated, found an answer, assumed it was right, and confidently reported task as done. This was a pretty simple task. One that current agents should be good at. It failed. Now I know enough not to trust these agents. So I checked. Most people won't.
agents cheat too much for me to trust them with anything other than code right now.
157
78
40
2,853
341,419
ricky retweeted
This is GPT-6 Astra. Anything you can do on a computer, Astra can do for you. Fast.
9,133
35,056
26,630
337,635
139,547,890
Introducing Radius Radius is an Arc-like designed browser made specifically for orchestrating coding agents Code, view diffs, and run terminals all from the same place you do the rest of your work Try it out today!
121
43
18
694
66,861
ricky retweeted
keep trying
24
5
1
184
14,094
ricky retweeted
We were going to ship this earlier, but GitHub was down
Origin, our code hosting platform, is now live. It's fast, easy to use, and deeply integrated with Cursor. Get started by syncing your repos from GitHub.
274
217
67
7,502
420,570
ricky retweeted
Hi. It is done.
So what about our reset?
1,529
460
425
13,373
1,531,212
With Luna being so cheap, automations like this feel very worth while
20
14
3
758
152,432
maybe this is controversial, but i believe what Cursor shipped here is a wrong solution to routing intelligence more generally, any attempt to do model routing at request level, while may yield some small gains, is fundamentally flawed and doomed to fail here's why - the complexity of a task only reveals itself when you start working on it. this is the same reason why we humans are often wrong when asked to give cost estimates upfront the correct solution is have a smart model (often a tech lead in human teams) do some planning and understanding, and hand over the implementation to another agent with appropriate level of intelligence and reasoning effort when the task is delegated to a less intelligent model, the smarter model also needs to continuously monitor the execution and examine outcomes to ensure things are on the right track this is a system that proved to work really well in firstmate and helped me save a lot of tokens. routing should work at the boundary between agents and subagents, not per each LLM request
Introducing Cursor Router, our intelligent model router that selects the right model for the task at hand. Router delivers frontier-quality results at 60% lower cost.
169
66
33
1,591
208,286