Strong opinions on AI tooling, software dev, and running a small profitable software business // Co-owner & software engineer at Sanico // Adelaide, Australia

Adelaide 🇦🇺
This is like the equivalent of discriminating against the user agent of the web browser: people end up forging because there is no other choice. In this case lucky figma sucks as a product so it’s not a big deal though.
so the choices are: - don't play by the rules, impersonate a sanctioned figma mcp client - play by the rules, click on the link to the suggested form and find out that you can't play by the rules if you want figma mcp server access for your users
2
70
I know of very few companies that do this. I am still holding off buying a new pc but when I do framework is the company for me.
We were able to get additional shipments of memory at this cost and have updated all Framework Laptop 13 Pro pre-orders configured with 32GB or 64GB LPCAMM2 to the new lower pricing. This pricing and also the X9 CPU config are available for new pre-orders too.
15
I’d argue this has been the case for quite a while. Over a decade ago I wanted to leave my university degree after 1 year because I had self-taught myself programming and compsci skills to a higher level than the university could ever give me. I only stayed because I knew several top tier companies at the time required those degrees to hire. I stayed for the certificate, not to learn the skills.
Most people expect that university degrees will lose value with the arrival of AI. It is likely to push more people to go for graduate degrees. That might be unhelpful: the traditional graduate thesis can be AI generated with relative ease. Everything else being equal, I think that graduate degrees will lose relatively more value. I was chatting with one of our graduates who is struggling a bit. I repeated the advice I gave several years ago when he graduated: « do something good and show the world... can be anything at all... but it must show competence. » His argument was that this is what he thought a university degree was. But that's absurd. It is like getting a degree in music and calling yourself a musician. Or getting a degree in creative writing and calling yourself a writer. It is what I call the 'paint by numbers' problem. You have books with a drawing and distinct sections, all with a number. You are expected to fill the regions with the designated colour. And then you have a painting! The problem is that you were 'on rails'. You did not really do the painting. And the same applies to all other university degrees. And AI is not helping at all because you followed obsolete rails. It is no longer very relevant. You studied latin for a world that works in English. It is true that in 1970, if you had a university degree, you'd probably be hunted down and hired. But 1970 is a long time ago. It becomes more and more pressing to work on actual skills. I am not advocating that you skip university necessarily. I am advocating for 'real projects' and 'real work' as early as possible.
1
24
Sav Tripodi retweeted
Good programmers state they use models like GPT 6 Astra without obtaining good results: it makes me feel I live in a parallel universe. But the explanation is not that hard: good programming in the past and now requires a different skill set, even if there is some overlap.
148
72
1,468
130,914
Never let a cool technology mask the factor of skin in the game.
Took a self-driving cab in Austin. So ridiculously cool. It had been a while since I’d felt like there are countries more technologically advanced than where I live. I hope Europe lets this happen soon too, now that the bottle cap emergency is under control.
8
Sav Tripodi retweeted
The wildly opposite ai programming takes are a sign the folks calling it over are missing what's possible, because they were limited by what they came from. Dream bigger! others: "it's over, guess it's all rust from here" me: "omg there's so many things I can build in Elixir"
13
25
183
21,149
I see you and raise you. Australia completely banned nuclear plants. Our power grid in South Australia (my state) sucks and we pay some of the highest prices in the world. And the worse part, ironically we have the highest concentration of uranium in the world. Beat that!!
Denmark wasted forty years on anti-nuclear nonsense, and now the grid is so fucked by variable wind supply that new commercial connections have to wait up to TEN YEARS. Tragic self-harm.
2
58
Sav Tripodi retweeted
(🌶️ take) You need better engineers to work with LLMs
34
19
452
23,647
Now this is what the future will be, precisely trained models, not the catch all giants currently being released. General purpose will always have its place but specifically trained for a specific problem just makes sense.
Training tiny models for special purpose use cases works so incredibly well if you have a great self improving recursive flywheel. Shopify ML team is on fire. finetuned 0.8b model beats GPT 5.6-sol xhigh in this very specialized task.
103
Finally, someone who gets via negative.
too many people working on making models smarter when they should be working on making models less stupid
54
People on x write some romantic sounding posts that says business is as simple as <INSERT TOPIC>. They’ll say project management, technical skills, or some other point they’re trying to push. Not sure if it’s just to say something online or to get a bunch of views but the reality is things aren’t that simple. To not do badly at business takes a bunch of skills they don’t teach you in any “school”. You ain’t gonna learn it in a 175 character tweet. Learn by experience not from some snake oil sales person.
43
Sav Tripodi retweeted
Omarchy is blowing up. I've never been involved with anything in my career that has grown this fast. With Ruby on Rails, we had years to build solid institutions, teams, and relationships. With Omarchy, we've been forced to figure it all out in twelve days. It's exhausting, but also incredibly exciting. It was never sustainable with just Ryan and me running everything. That's how it was, more or less, up until Quattro. Lots of other contributors, but all the responsibility was on us to make sure the ship stayed afloat, the servers didn't crash, and fixes got pushed out quickly. Now it's time to build a proper institution. Durable, resilient, and competent. That's what we're doing here on Basecamp now. I'm spinning up teams for every facet of responsible distro management, and I'm getting an absolute outpouring of interest for all of them. Everyone wants to be part of this. We're winning hearts, minds, and volunteers at an astounding rate. Great! We need all of it to succeed. Because make no mistake: There are many people who'd love to see this rocket blow up before it reaches the moon. Aggrieved Linux users who don't like the sudden attention their exclusive hobby has received. Competing Linux distributions that are seeing our numbers explode. Mac stans who've sunk their identity into an apple. And, of course, any of the haters I've picked up in my quarter-century career speaking bluntly on the internet. They're not going to succeed. Because we've already become unstoppable. There's too much momentum, too much money, and too much support now backing this effort. We're living the Mandate From Heaven meme at the moment. And we are here to fulfill the prophecy: The Year of Linux on the Desktop! That has been a joke for two decades. But by the end of the year, nobody at Apple or Microsoft is going to be laughing. They're going to be scrambling. Because neither of these proud organizations currently has any method to counter the speed, vision, or ambition with which we're going to accelerate into the future of personal computing. This is the moment. This is the opening. This is our chance. For thirty years, we've been subject to one OS overlord or another. Dictating how we compute. Choking off competitors through platform malfeasance. Tollboothing the distribution. That ends now. Because Linux is going to win. And Linux is free. As in beer, speech, and source. But just because it's inevitable doesn't mean it's going to be easy. We have a lot of work in front of us if we actually want to make our mark. But there's never been a better time for this kind of delusional ambition. The age of agents is the unlocking factor. It sounds like a LinkedIn slogan, but it's true. Where the application of tokens goes, the innovation follows. We can fix everything. Let's do it together. Let's go. --- This is what I sent to the dozens of new volunteers who've signed up for teams within the new Omarchy organization yesterday. But we might as well broadcast our mission and intentions to the world too.
360
566
8,937
901,007
Far, far down the list to building a beautiful family, the worlds most invaluable asset.
A startup making money while you sleep is still one of the coolest things you can build in life.
41
Soon even the local supermarkets will be releasing their frontier AI models 🤦‍♂️
Meet Granite 4.2, IBM’s latest family of open models purpose-built for enterprise agentic AI. With new native reasoning capabilities, Granite 4.2 can plan, reason, self-correct and reliably use tools to help automate complex enterprise workflows ⬇️
75
Sounds like a bunch of follow up "explaining", while evading the real reason claude didn't respect the standard after all this time. Just fix it without the "explaining" part.
Replying to @tobi
Hi Tobi, thanks for the feedback! We are working on making Claude Code more hackable, which will include being able to easily use Agents.MD or make other system prompt modifications. I’ll share more here when it’s ready to roll out. Our reasoning for this was that we don't think that model families are interchangeable and the system prompt can have a big impact on performance. I wrote here: x.lingyaoai.com/trq212/status/20807109… on how Claude models like their skills, system prompt, Claude.MD, etc to be formatted. In Claude Code we have different system prompts per model. That said, I know this is a lot of upkeep and not worth the squeeze for everyone. Feedback heard and I'll keep you posted. In the immediate term, you can always @Agents.MD from your Claude.MD.
110
Love the enthusiasm but I’m convinced you need the haters to an extent. It helps fuel the pathway to proving them wrong. Also, if people are crusading against it then it means he’s on the right track. It will also inspire the haters to create better alternatives. We all win!!
I have never seen a take trigger so many retards into hallucinating and blasting out the worst possible criticisms. Yeah, he's rich and maybe he's a little hypocritical to make political points while saying he's being apolitical, but like. Who the fuck cares. He's bringing people into open source, and now the screeching retards once again make the normies go "wow I want nothing to do with this". SHUT THE FUCK UP AND EMBRACE CHOICES AND GROWTH IN THE OPEN SOURCE ECOSYSTEM. Remember all of the times some rich and powerful entity did some things and it made FOSS better in the end because we can always fork projects that get hijacked? Look at who is paying the Linux Foundation and other open source institutions, and look at the people who are actually putting in the work to build out open source to make the entire ecosystem better. They're not random brokies. If you don't like it, shut the fuck up. Fork. Write your own code. Maintain your own shit with 500 million users for free. I don't care. If you're not going to do this, shut the fuck up and get the fuck out of MY industry. Fuck you.
2
1
139
Influence can go a long way, many of my clients end up in circles with support teams only to get denied by a generic response like the one from spaceship. @dhh isn’t justin bieber famous but has significant leverage in the tech industry. He will make linux on the desktop happen.
Replying to @dhh @spaceship
Hey DHH, I'll get someone to take care of this asap.
63
Why can @tobi threaten this? Readily alternative options!!! The difference and capability of the current frontier models isn’t huge. A year ago nobody could say this. Maybe that’s why most AI companies are pushing for harnesses? Everyone is scrambling for a competitive edge.
I’m thinking about banning Claude code at Shopify until they change their mind and read AGENTS.md and .agents/skills etc. Insisting on only reading CLAUDE.md sometimes leads to split brain problems when different team members use different tools. Just unnecessary.
1
30
That’s why I love what @TimSweeneyEpic is doing with apple and google. Readily alternatives means you can change vendor if you dislike their behaviour, want better service, or a more affordable option. The current AI landscape allows for this, let’s hope it stays that way.
19