Profile avatar
narphorium.bsky.social
Building AI powered tools to augment human creativity and problem solving in San Francisco. Previously @GitHub Copilot, @Google, 🇨🇦
166 posts 1,279 followers 553 following
Regular Contributor
Active Commenter
Posts 23 Comments 27

Claude 4 vs Claude 3.7 is like when you're in a meeting and you can tell who read the doc beforehand vs who's skimming it live just based on the questions they ask

submitted 15 hours ago • 1 comment

“When a senior engineer joins a team and opens a large monorepo, they don’t read isolated code snippets. They scan folder structure, explore files, look at imports, read more files. That’s the mental model your coding agents need to replicate…” open.substack.com/pub/pashpash...

submitted 1 day ago • 0 comments

I like how they just let Dario go out on stage and be Dario instead of coaching him to do the Steve Jobs imitation act that so many tech CEOs do these day www.youtube.com/live/EvtPBaa...

submitted 2 days ago • 0 comments

"SEATTLE, May 20 (Reuters) - Microsoft (MSFT.O), said it will add an artificial intelligence coding agent from Google-backed startup Anthropic to its GitHub service." You have to have worked at at least 2 out of 3 of these companies to even parse this sentence 😂

submitted 4 days ago • 0 comments

Will we see a generation of kids raised on AI video slop because, to adults, it’s indistinguishable from Cocomelon?

submitted 4 days ago • 0 comments

@orionreed.bsky.social and I have been working on a custom HTML attribute that binds a language server (LSP) to a input/textarea/contenteditable, which adds functionality like diagnostics, autocomplete, semantic renaming, etc.

submitted 16 days ago • 3 comments

Making progress on my relational layout lib. Did a pass on improving the API and hopefully the code is more readable now.

submitted 16 days ago • 2 comments

This should be a really fun talk, Shawn Simister and I have been recalling to ourselves how it all worked. What we realized is that we were doing an early form of LLM-as-judge back in early 2023. maven.com/p/da8264/how...

submitted 17 days ago • 0 comments

Twitter has become a broadcast medium, which is disorienting to those of us who remember it as a communication medium

submitted 17 days ago • 0 comments

"Automatically setting up, building, and testing a developer environment to aid Agentic Software Development." with tamás and davidslater githubnext.com/projects/dis...

submitted 18 days ago • 0 comments

Voice editing with change review in @obsidian.md 🌴 youtube.com/shorts/tB7mc...

submitted 19 days ago • 1 comment

Claude Code now supports resuming sessions, which means that session logs (including tool/llm calls) are stored locally. My curiosity got to me, I figured out the format and wrote a CLI tool to dump specific conversations github.com/paulgb/claud...

submitted 23 days ago • 0 comments

Unpopular opinion but this season of Andor feels rushed.

submitted 25 days ago • 1 comment

This is an interesting look at the traceable text design pattern by @hitakam.bsky.social and @andrewhead.bsky.social dl.acm.org/doi/10.1145/...

submitted 28 days ago • 1 comment

At #chi2025 this year, see our group's latest: a projectional editor for augmenting math formulas, and a study of augmented AI-generated texts. Look out for these faces! Jeff, @hitakam.bsky.social, and I are excited to reconnect with old friends and make new ones.

submitted 28 days ago • 1 comment

Finally found the time to migrate my research notes off of LogSeq to @obsidian.md . I still prefer graph-shaped notes vs files and folders but LogSeq sync was driving me crazy

submitted 30 days ago • 0 comments

what if you could model a problem in a spreadsheet, and then instantly start applying powerful *filters* to explore your space of options? that's the idea @alexwarth.bsky.social and I have been exploring recently at @inkandswitch.com ... 1/

submitted 32 days ago • 4 comments

In our recent piece for @noemamag.com, my colleague James Manyika and I share five fundamental and interconnected paradigm shifts in our understanding of intelligence. Over the coming days, I’ll dive into each. First, natural computation.

submitted 36 days ago • 1 comment

Nubya Garcia at the Chapel. One of the best performances I've experienced in a while

submitted 37 days ago • 0 comments

I still remember how excited I was when I got access to the first Codex model github.com/openai/codex

submitted 38 days ago • 0 comments

I think many people don't use custom cursor rules because they don't know where to begin. I'm experimenting with a set of "metarules" which I can use in chat to generate custom rules for any project

submitted 39 days ago • 0 comments

www.geoffreylitt.com/2025/04/12/h...

submitted 41 days ago • 7 comments

I've started building a better Cursor rules editor which analyzes my chat history to automatically suggest improvements or remove unused rules. 🧵 1/6

submitted 45 days ago • 1 comment