bgirard

5 ms·

31.
by bgirard 8mo ago
About ~$300: $200 for Claude max subscription $20 for Vercel $20 for Codex $20 for Meshy I think these days the $200 Max subscription wouldn't be needed. I bet with these latest models you can make due with mixing two $20/mo subsc
32.
by bgirard 8mo ago
Thank you. There's a demo save to get the full feel of it quickly. There's also a 2D-ASCII and 3D render you can hotswap between. The 3D models are generated with Meshy. The entire game is 'AI slop'. I intentionally did
33.
by bgirard 8mo ago
Doesn't feel like a useful data point without more context. For some hard bugs I'd be thrilled to wait 30 minutes for a fix, for a trivial CSS fix not so much. I've spent weeks+ of my career fix single bugs. Context is everyt
34.
by bgirard 8mo ago
> Using the develop web game skill and preselected, generic follow-up prompts like "fix the bug" or "improve the game", GPT‑5.3-Codex iterated on the games autonomously over millions of tokens. I wish they would share
35.
by bgirard 8mo ago
The switching cost is so low that I find it's easier and better value to have two $20/mo subscription from different providers than a $200/mo subscription with the frontier model of the month. Reliability and model diversity
36.
by bgirard 9mo ago
> malicious It doesn't have to be malicious. If my workflow is to send a prompt once and hopefully accept the result, then degradation matters a lot. If degradation is causing me to silently get worse code output on some of my commi
37.
by bgirard 9mo ago
Revenue per corporate employee are both ~1.5m/corp employee. I think it really comes down to how the company is managed and culture.
38.
by bgirard 9mo ago
When you have a personality disorder like NPD, you'll believe to your core that every criticism of you is a lie. When you're in an abusive relationship they say intentions don't matter, only impact does. Because victims often
39.
by bgirard 9mo ago
~25 years isn't the average when you account for accidents, rust, and useful economic life of a car. We had 200+ car crashes due to weather in a single day this week in Waterloo, Ontario.
40.
by bgirard 9mo ago
It's not from a few prompts, you're right. But if you layer on some follow-up prompts to add proper test suits, run some QA, etc... then the quality gets better. I predict in 2026 we're going to see agents get better at runni
41.
by bgirard 9mo ago
Yea. When my girlfriend and I travel I use Google Maps and she uses Apple Maps. And it's a running joke at this point that if we're struggling to get somewhere, she'll pull up Apple Maps and get us unstuck and bug me to switc
42.
by bgirard 10mo ago
Wow, crazy to see this childhood classic load and run seamlessly in my browser. Better than on my old hot wheels pc.
43.
by bgirard 10mo ago
I used to think this way. But I moved to AGENTS.md. Now I use the different UI as a mental context separation. Codex is working on Feature A, Gemini on feature B, Claude on Feature C. It has become a feature.
44.
by bgirard 10mo ago
Why wouldn't you switch? The cost to switch is near zero for me. Some tools have built in model selectors. Direct CLI/IDE plug-ins practically the same UI.
45.
by bgirard 10mo ago
What's your definition of a novel idea? How do you measure that? I've had a 15 year+ successful career as a SWE so far. I don't think I've had a single idea so novel that today's LLM could not have come up with it.
46.
by bgirard 10mo ago
This includes people on a waitlist for hip operations. Are those life-threatening?
47.
by bgirard 10mo ago
Right. It finds an economic optimum limited to what people are willing to pay and their perceived value from that. That optimum doesn't naturally weight more complex factors like for the trade-off for the smog generated from your trave
48.
by bgirard 10mo ago
That's a straw man. I didn't say personal vehicles.
49.
by bgirard 10mo ago
Not surprising. The real question is how do we measure the opportunity cost of these measures? Is it a net gain? You could, at the extreme, ban all motor vehicles but the opportunity cost would outweigh the benefits.
50.
by bgirard 11mo ago
It's a fun trope to repeat but that's not what OpenAI is doing. I get a ton of value from ChatGPT and Codex from my subscription. As long as the inference is not done at a lost this analogy doesn't hold. They're not payi
51.
by bgirard 11mo ago
This is one of many frequent reminders: In some environments, how you brand and market your work (Mush with a chainsaw cutting spending comes to mind) is often more important than the work you do. Most wont bother to look at the actual resu
52.
by bgirard 11mo ago
Is there anything more concrete than that? Large wins on their own aren't meaningful if they aren't good risk adjusted trades or repeatable. I've made big wins but I don't consider myself a good trader.
53.
by bgirard 11mo ago
I'm legitimately curious what his overall performance is. Are those just cherry picked trades?
54.
by bgirard 1y ago
> Another was a code that was inpatient only and because it was an emergency he had never been admitted. The threads says this was 4 hours of work and they billed for things that weren't even used.
55.
by bgirard 1y ago
I used AI to deal with customer support when a company tried to assign me the rental contract from the previous owner. ChatGPT correctly quoted the relevant Ontario Consumer Protection Act sections that applied. I just did quick verificatio
56.
by bgirard 1y ago
That makes sense. But then the idea of a simultaneous failure preventing, say AWS and GCP/Azure, from restarting due to accidental circular dependencies in the init sequence needed to reboot the two competing cloud infrastructure is in
57.
by bgirard 1y ago
The big issue I have with this experience is that you don't get a clear charge price before you leave. So you have to check a page either some minutes or hours later and hope that the total is correct. Like the article said, I don&#x27
58.
by bgirard 1y ago
> The person in front of me bought two items and saw she got charged for three. Since there were no paper receipts, she took a photo of the machine before going to the guest services to complain. I missed ten minutes of the game getting
59.
by bgirard 1y ago
I have a very similar experience. I was heavily invested in Anthropic/Claude Code, and even after Sonnet 4.5, I'm finding that Codex is performing much better for my game development project.
60.
by bgirard 1y ago
Looks promising. I got my environment working well with Codex's Cloud Task. Trying to same repo with Claude Code Web (which started off with Claude Code CLI mind you), and the yarn install just hangs with no debuggable output.

More ›