Michał Czyż (@cs3b) header
Michał Czyż (@cs3b) profile picture
Arona, Spain
Joined March 2010
Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)Photo by Michał Czyż (@cs3b)
Michał Czyż retweeted
Just passed 10,000 followers. Thank you. You shouldn't need a $15K rig to run a serious local AI model. I want everyone running the most powerful model their budget allows, starting with the hardware they already own. This week: - A 2021 M1 Max got 25-29% faster GPU decode on Linux from one power-management fix - GPT-OSS 120B on a laptop plus an RTX 5060 Ti eGPU, 25 tok/s - Qwen3.8 Flash Next at 89 tok/s on a DGX Spark - Omarchy on a Mac mini M6, first GPU llama.cpp run included My AI hardware has outgrown my home office, as you can see! I'm turning our spare room into an AI lab. Every machine I can test becomes a budget tier I can show you how to use. The lab still needs two 100G cables and a Thunderbolt 5 PCIe box so the ConnectX-5 isn't stuck at Thunderbolt 3 speed. If you want to help, sponsor me or send hardware: joshuawarren.com/sponsors I'll post what arrives and what it unlocks. And as always, all of my work is open source, on my GitHub.
56
24
7
460
21,967
Michał Czyż retweeted
Day 25/ We are doubling all usage limits for all Pro 100/200/500 tiers and Business Premium, effective today. We heard your kind feedback and won’t be reducing usage limits for Pro 200, so your Work and Codex usage will stay the same. We’re also going to continue offering extended limits for dots for now, and usage of your dot will still continue to be separate from your default usage. And we'll of course also keep shipping better models and other efficiency improvements that increase your effective usage over time.
7,363
2,699
3,648
39,115
7,585,638
Michał Czyż retweeted
Tibo已经没招了,我来出几招 1. plus取消5h限制 2. dot给plus用户使用 3. ultrafast改为提升15倍速度 4. Pro20x的用量翻倍 5. 提升6.1 Sol的TPS至少到50 这已经够Tibo 5天的了🤓
Either we are out of ideas or we are cooking. Time will tell
163
47
18
3,157
455,127
Michał Czyż retweeted
My harness tier based on my own usage. Every single one I have used bar anti-gravity
52
32
7
974
95,206
Michał Czyż retweeted
oh no - codex $200 subscription's dollar value is being cut in half in their defense, openai's models are indeed quite efficient. but anthropic has caught up too and the big cut is only the first order effect the second order effect is that this will give anthropic enough room to cut their subscription value as well the third order effect is that the labs have found a way to gradually end the subsidization for consumers subscriptions. just do this a few more times, and we'll see the subscriptions charging the same as API pricing i predict that in about a year, we'll refer to what we had as the good old days
Hi, Tomorrow we are re-opening the Pro $200 subscriptions to new subscribers, but together with it we are also changing how we calculate the usage for it. In effect, if you do the math, it will net out at half the dollar in API spend compared to the old Pro $200 plan. Now that it's said, let me explain why this is happening and why you will still get more work done than if you were on the Pro $200 subscription one month ago. (a) We didn't want to compromise in other ways and are committing to not reintroducing the 5h limit, so that you can fully use the weekly usage when you want. (b) On the subscription, we guarantee that over time you always get more work done and with an increasing level of quality. This means that you will continue to get more value per dollar spent as a result of models getting more efficient and us passing down the improvements in the form of API price reductions. (c) We don't want to put an incentive on ourselves to artificially inflate the API list prices to make it look like you are getting a lot (and workaround it through discounts, etc). Instead we want to continue to both rapidly reduce prices and increase capabilities of models on the API. This week we introduced GPT-6 Sol and GPT-6 Luna at 50% of their previous price. Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent. (d) Tomorrow, we are adding more things to the subscription that won't draw on the usage, I won't reveal what that is yet. I wanted to be transparent before all the big announcements tomorrow. Lots of new exciting things are coming to the subscriptions that will make it super compelling, but I wanted to make sure to share this change ahead of time so you can all understand it before we shower you with good news. Codexingly, Tibo
136
27
16
903
110,254
Michał Czyż retweeted
Prediction time. @dhh just asked over a thousand attendees at #railsworld how many people still write code by hand on a regular basis. Less than 10 people raised their hand. Let that sink in. Next year the question will be how many people are still doing manual PR code reviews or directly looking at code on any regular basis. Result will be similar.
47
68
14
628
69,815
Michał Czyż retweeted
Top Agentic Development Environments for 2026 #1 - Zed - Open Source - ACP support - Claude without their god awful harness - Project tabs - Performant - SSH tabs - Beautiful #2 - T3Code - Open Source - Good, consistent team - Claude without the god awful harness - Performant - Looks Fine #3 - CMUX - Ghostty terminal works amazing w all agents - Nice browser feature altho I don't like webkit - MORE PANES BRO MORE PANES - Tabbed projects #4 - Ghostty - Bare bones - Most performant #5 - Codex App - Feels nice I use this the most of all the "apps" - I love automations and plugins - Needs to be open sourced
76
20
3
538
64,936
Michał Czyż retweeted
It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow. Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes. As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now. It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.
1,589
4,639
1,113
36,760
5,180,018
Michał Czyż retweeted
I’m joining OpenAI Codex to work on the future of agentic development! At Cursor, I got to see the shift from autocomplete to agents. The next step isn’t a better IDE. It’s an Agent Development Environment (ADE): systems and tools for orchestrating agents, reasoning over their outputs, and making them autonomous enough to reliably complete ambitious work. After chatting with @embirico and @thsottiaux, it was clear that Codex is the best place to realize this vision. The team has consistently shipped SOTA models for agentic coding (check out gpt-5.3-codex) and I’m pumped for the future that the new Codex App points to. What I’m most excited about is the broader mission: accelerating the knowledge work economy. All agents are coding agents, and we’re already seeing Codex used across every job function within organizations. I’m extremely grateful for my time at Cursor, working with the incredible team, and I’m proud of what we built together. I’m excited to take an even bigger swing with Codex. If you’re curious to get a glimpse of where we are headed, download the Codex App! If you want to work on this mission, please apply or reach out - we are hiring across all functions! You can just build things.
214
98
29
2,666
760,573
Michał Czyż retweeted
damn I should stop using markdown files in git
40
4
1
400
122,452
Michał Czyż retweeted
millennial gamers are the best prepared generation for agentic work, they've been training for 25 years
269
622
176
7,324
1,532,325
Michał Czyż retweeted
"I also get some programmers are eager to tune it out. The hype drones on, the fantastical claims are still far off, and there's uncertainty where this leaves the profession. But that's not reason to miss out on this incredible moment in human history!" world.hey.com/dhh/promoting-…
59
89
25
1,012
152,546
Michał Czyż retweeted
Using a mocap suit to kick yourself in the balls with a robot is a great metaphor to close out 2025.
644
15,491
1,189
126,033
5,883,632
Last quarter I rolled out Microsoft Copilot to 4,000 employees. $30 per seat per month. $1.4 million annually. I called it "digital transformation." The board loved that phrase. They approved it in eleven minutes. No one asked what it would actually do. Including me. I told everyone it would "10x productivity." That's not a real number. But it sounds like one. HR asked how we'd measure the 10x. I said we'd "leverage analytics dashboards." They stopped asking. Three months later I checked the usage reports. 47 people had opened it. 12 had used it more than once. One of them was me. I used it to summarize an email I could have read in 30 seconds. It took 45 seconds. Plus the time it took to fix the hallucinations. But I called it a "pilot success." Success means the pilot didn't visibly fail. The CFO asked about ROI. I showed him a graph. The graph went up and to the right. It measured "AI enablement." I made that metric up. He nodded approvingly. We're "AI-enabled" now. I don't know what that means. But it's in our investor deck. A senior developer asked why we didn't use Claude or ChatGPT. I said we needed "enterprise-grade security." He asked what that meant. I said "compliance." He asked which compliance. I said "all of them." He looked skeptical. I scheduled him for a "career development conversation." He stopped asking questions. Microsoft sent a case study team. They wanted to feature us as a success story. I told them we "saved 40,000 hours." I calculated that number by multiplying employees by a number I made up. They didn't verify it. They never do. Now we're on Microsoft's website. "Global enterprise achieves 40,000 hours of productivity gains with Copilot." The CEO shared it on LinkedIn. He got 3,000 likes. He's never used Copilot. None of the executives have. We have an exemption. "Strategic focus requires minimal digital distraction." I wrote that policy. The licenses renew next month. I'm requesting an expansion. 5,000 more seats. We haven't used the first 4,000. But this time we'll "drive adoption." Adoption means mandatory training. Training means a 45-minute webinar no one watches. But completion will be tracked. Completion is a metric. Metrics go in dashboards. Dashboards go in board presentations. Board presentations get me promoted. I'll be SVP by Q3. I still don't know what Copilot does. But I know what it's for. It's for showing we're "investing in AI." Investment means spending. Spending means commitment. Commitment means we're serious about the future. The future is whatever I say it is. As long as the graph goes up and to the right.
5,082
25,163
5,922
169,564
26,309,294
Michał Czyż retweeted
now at 92% weekly limit, 2 days into the week anthropic told us that only "2% of users" would be affected by this it has now become apparent that the "2% of users" referenced here meant "people trying to build stuff with claude". they dont want you to do that anymore
used 86% of claude's weekly limit on the $20 plan in less than 2 days on sonnet 4.5. each single session is about 10% never hit the weekly limit before using opus 4 or 4.1
59
24
4
490
67,677
This is real Claude Code hits usage limits for the week in hours. I've never once hit their 5hr limit on the 20x plan I think this is the most egregious usage limit change thus far in the industry
248
85
51
1,549
282,668
Michał Czyż retweeted
A lot of negative chatter about Claude Code usage limits being hit very quickly Enough that Anthropic are now looking into it…
104
21
11
588
116,749
Michał Czyż retweeted
I'm constantly starting tests and demos of Rails apps and other tools. This new try tool from @tobi looks exactly what I didn't even realize that I needed! github.com/tobi/try
33
35
7
638
69,468
"Hey Andrzej, can you help me get into Ruby?" I was asked this question 4 times in the last 2 weeks only. All by >10yrs experienced programmers from Java, .NET, Python, JavaScript. Why? Because they all watched the recent @dhh interview with @lexfridman and they want to be part of the Ruby luxury world too.
1
7
78
4,335