Lightnews — Scholar-powered news

Shawn Simister

@narphorium.com

1.3K followers 600 following 240 posts

Building AI powered tools to augment human creativity and problem solving in San Francisco. Previously @GitHub Copilot, @Google, 🇨🇦 narphorium.com

Posts Media Videos Starter Packs

Pinned

Shawn Simister @narphorium.com · Aug 6

I wrote up some thoughts about why we’re seeing a gap in AI tool use among different types of developers
narphorium.com/blog/top-dow...
1/4

Top-Down vs. Bottom-Up Development | Shawn Simister

Why do AI tools work for some developers but not others?

narphorium.com

3 1 5

Reposted by Shawn Simister

Rahul Pandita @rahulpandita.bsky.social · 1h

Some thoughts on LLMs, Autopilot, and VibeCoding rahulpandita.me/blog/2025-10...

1 1

Shawn Simister @narphorium.com · 2h

These new Windsurf codemaps look a lot like Project Copernicus from @kcieslak.io
githubnext.com/projects/cop...

Reposted by Shawn Simister

Sebastian Raschka (rasbt) @sebastianraschka.com · 3d

6/
That said, HRM and TRM are fascinating proof‑of‑concepts that show what’s possible with relatively small and efficient architectures. I'm still curious what the real‑world use case will look like. Maybe they could serve as reasoning or planning modules within a larger tool‑calling system.

2 2 9

Shawn Simister @narphorium.com · 5d

Its still under development so I don't know exactly how but they seem to have some way of reasoning about which parts of the code satisfy which functional requirements

Reposted by Shawn Simister

Sebastian Huckleberry @sourishkrout.com · 5d

We have 🤖 AI notetakers in meetings but continue to silo know-how every time we close terminals. Not just the how but also the why and what.

Sign up for beta access to visr.sh: it's like as if Granola AI and #tmux had a kid.

Visr: The Agentic Terminal Notepad Integral to Your Docs

The Agentic Terminal Notepad Integral to Your Docs

visr.sh

2 5

Shawn Simister @narphorium.com · 6d

Fun panel discussion on spec-based development with @kiro.dev team.
Especially interested in upcoming features to ground the specs to verify that the work is done correctly

1 1

Reposted by Shawn Simister

ben ✶ @bf.wtf · 9d

What, if anything, do LLMs change about interface design?

Interfaces are languages

Is there a universal translation?

shimmeringvoid.leaflet.pub

2 8

Shawn Simister @narphorium.com · 9d

The Trailblazer code is open source so you can try the tool in your favorite VS Code fork: github.com/YanLitao/Tra...

Shawn Simister @narphorium.com · 9d

As autonomous agents tackle larger tasks over longer trajectories, we’ll need tools like Trailblazer to quickly walk us through how problems were solved so verification doesn’t take as long as solving it ourselve

Andrew Head @andrewhead.bsky.social · 13d

Today at #uist2025 my student Jeffrey Tao presented our new paper "Answering Developer Questions with Annotated Agent-Discovered Program Traces" While it is too late to catch the talk, you can still find Jeff! I'm happy to make connections over email.

Paper: andrewhead.info/assets/pdf/t...

1 3

Shawn Simister @narphorium.com · 15d

Teaching LLMs to Plan: Logical Chain-of-Thought Instruction Tuning for Symbolic Planning
arxiv.org/abs/2509.13351

2 8

Shawn Simister @narphorium.com · 19d

Yes! Same. I think of them as stepping stones narphorium.com/blog/top-dow...

Top-Down vs. Bottom-Up Development | Shawn Simister

Why do AI tools work for some developers but not others?

narphorium.com

Reposted by Shawn Simister

Gregor Samosa @watwa.re · 25d

My ideas are faster than my fingers so I've did some Figma prototyping and video editing to talk about some of my new ideas before they are actual coded reality
youtu.be/2DRWUgK-RdA

hirba - Legible live programming through rich in-context representations

YouTube video by watware

www.youtube.com

1 6

Shawn Simister @narphorium.com · 25d

I once booked an appointment at the genius bar only for them to pull that fluff out and hand my phone back to me. Embarrassing

1 2

Reposted by Shawn Simister

Alonso Silva @alonsosilva.bsky.social · 26d

Want to visualize the response format constraints on the LLM when working in a Jupyter notebook?
Then you might be interested in my new project `litelines`.
Litelines lets you visualize the selected path by the LLM.
It supports a Pydantic schema as a response format, as well as regular expressions.

1 4 8

Shawn Simister @narphorium.com · 28d

Sometimes you need to build the wrong thing to get in the right headspace to build the right thing

Reposted by Shawn Simister

𝕮 @chrisshank.com · Mar 1

"Generative, Malleable, and Personal User Interfaces" by Haijun Xia (2025)

www.youtube.com/watch?v=MbWg...

www.youtube.com

1 6

Reposted by Shawn Simister

Oskar @austegard.com · Sep 10

Advanced context continuity with Projects and the Claude Pruner bookmarklet whtwnd.com/austegard.co...

Managing Context Continuity in Claude Projects | Oskar

Managing Context Continuity in Claude Projects The Problem You're deep in a productive Claude conversation. You've built up complex context, made breakthroughs, developed solutions. Then you hit the...

whtwnd.com

1 2

Reposted by Shawn Simister

grant @grantcuster.com · Sep 9

Need to rethink my gesture math and try some polarizing film, but basics of panning the canvas are in there.

1 1 6

Shawn Simister @narphorium.com · Sep 9

bsky.app/profile/narp...

Shawn Simister @narphorium.com · May 22

Thinking about AI alignment and LLM hallucinations and I remembered how as college students we were taught that it's better to make up an answer on a test and get partial marks than to leave an answer blank 🙃

Shawn Simister @narphorium.com · Sep 6

Cursor limits me to 5 custom modes so I reserve those slots for coding-specific tasks and do all my broader design thinking in Claude Desktop instead. Interesting how small constraints in our tools quietly redirect our entire creative process.

Shawn Simister @narphorium.com · Sep 4

Bottom-up development?

bsky.app/profile/narp...

Shawn Simister @narphorium.com · Aug 6

I wrote up some thoughts about why we’re seeing a gap in AI tool use among different types of developers
narphorium.com/blog/top-dow...
1/4

Top-Down vs. Bottom-Up Development | Shawn Simister

Why do AI tools work for some developers but not others?

narphorium.com

1 1

Reposted by Shawn Simister

David Gasquez @davidgasquez.com · Sep 4

Is there a name for the pattern where you try building the same thing (feature, project, ...) a bunch of times with different agents to get a sense of the implementation design space?

I usually run things at least 2 or 3 times until I "feel" what is the way I want it to be built.

4 1 5

Shawn Simister @narphorium.com · Sep 3

Software engineering is evolving into software research. AI can now recreate most known solutions, so human value shifts from "how to build it" to "what should we build next?" The future belongs to those who discover new problems worth solving

Shawn Simister @narphorium.com · Aug 30

It’s not just about choosing the right amount of context. It’s about arranging it to tell a compelling story which leads the reader to a deeper understanding of the code

Shawn Simister @narphorium.com · Aug 30

Love this analogy from the RepoAudit paper. LLMs with huge context windows are like putting all your code on a single PowerPoint slide and then projecting it onto a 1000-foot screen

arxiv.org/abs/2501.18160

1 1 5