Viewing profile — philipbjorge

Wow, I would love to read an interview series based on this!!

It’s been 25 years since I was in school and this was my experience. Unsure if it’s changed…

I’ll second this, I had to send this message… > based on our discussions so far, I think we’d only need... Which they appear to have but I can't dig in on mobile.

I'll go further and note that some of the optimizations I've seen in rtk for things like `git status` have actually bubbled up into the model layer -- Codex is regularly making too…

I can't find the relevant issues in their repo, but I've been somewhat skeptical of their tool over-reporting token savings and there are many issues to that effect in the repo. I'…

I've done some pretty incredible things with LLMs. If this were sqlite with its exhaustive test suite... OK, I can see it. It's hard for me to see this not becoming a pile of slop,…

Been loving pi and codex lately. Good to build resiliency and self sufficiency into these systems.

Can you fill me in on how this impacts conductor? How are they using `claude -p`?

Tracking with `ccusage`, I pretty easily hit $2000/mo in API equivalent credits and while I'd consider myself a power user, I'm a responsible one that's generally always in the loo…

This seems like less of a today thing and more of an ancient human tendency. A lot of Buddhist practice is basically trying to train against immediately collapsing reality into sel…

I haven't really shared what I use, I'm still deciding if that's something I want to do. To get an idea of what I'm talking about, you could install https://github.com/obra/superpo…

Ahh good point -- I've handled this by switching my harness to `pi` but recognize that may not be for everyone and doesn't directly address OP's question.

What I found was that I *strongly* preferred Claude Code with its defaults. Codex was almost unusable to me -- It would spit out a 4-5 page plan where it kept repeating itself, whe…

You might search for a concept like `/handoff` that's in ampcode. I'm sure someone's built a skill for just this.

So happy to have diversified my model providers this past couple of weeks. GPT-5.5 has had no trouble slotting into Opus workloads. Will be fun to try out more of the models as tim…

I’ve been comparing Claude Code and Codex extensively side by side over the past couple of weeks with my favorite prompting framework superpowers… From my perspective, Claude Code …

gpt 5.4 has been performing great in my harness.

This looks remarkably similar to https://github.com/vercel-labs/agent-browser How is it different?

Until it's happened to you, it sounds unbelievable Sorry about all the broken plastic on the trim -- That's also very familiar...

> We asked for a bill with the standard CPT codes. No reply. Asked again. “Oh, we meant to send it. We upgraded our computers five months ago and nothing works.” Uh-huh. Finally go…

We had a similar realization here at Thoughtful and pivoted towards code generation approaches as well. I know the authors of Skyvern are around here sometimes -- How do you think …

Important is subjective — In the healthcare space, I’d make the claim that most applications don’t expose themselves correctly (native or web). CV and direct mouse/kb interactions …

If you're looking to test an LLMs ability to solve a coding task without prior knowledge of the task at hand, I don't think their benchmark is super useful. If you care about under…

devcontainers extension was a year out of date up until the last month or something? sorry, this is from memory, but definitely not 100% compatibility.

I haven't followed this closely, but I assumed it was related to a foreign entity having the ability to hyper-target content towards said 17 year olds (and the entire userbase in g…