Jess Martin
LightNews LightNews
banner
jessmart.in
Jess Martin
@jessmart.in
1.7K followers 360 following 650 posts
🛠 Tools for a better tomorrow
🧠 Architect-philosopher
💼 Building https://worksquared.ai
🏠 https://jessmart.in
📍Fuquay-Varina, NC
Posts Replies Media Videos
jessmart.in
friggin' love Conductor.build
November 5, 2025 at 5:31 PM
jessmart.in
wikipedia-mcp already exists [1] but I find the agentic CLI harnesses to be a lot more token-efficient over local content. Perhaps you could do both - run Wikipedia locally + local MCP server.

Here are the tools in wikipedia-mcp

[1] github.com/Rudra-ravi/w...
November 4, 2025 at 8:56 PM
jessmart.in
/exit should be /kill

Just look at that cute lil face
October 31, 2025 at 4:55 PM
jessmart.in
Good times! Cool to see Claude Code one-shot a @livestore.dev Kanban app.

I have a pretty good idea of the overall architecture of Claude Agents SDK, and I'll be able to get into the meat of it during the next livestream.
October 3, 2025 at 4:04 PM
jessmart.in
Bless you, Conductor
September 4, 2025 at 1:01 PM
jessmart.in
I've made a *lot* of @tldraw.com's over the years, but these two might be my most favorite, ever

#nocontextwhiteboards
August 20, 2025 at 3:35 PM
jessmart.in
Been thinking a lot about "building my software factory."

I have a pretty solid process and can keep up with two Claudes running in parallel, but I need to build some specialized agents:
- code review (just first pass, I still review)
- design work (claude can't match a visual spec yet)
July 23, 2025 at 4:53 PM
jessmart.in
Engelbart's C Activities
July 17, 2025 at 1:50 PM
jessmart.in
Here's the `/docs` directory for a project I'm coding with Claude Code

really helps having all of this context inside the project
June 26, 2025 at 11:38 AM
jessmart.in
here were some of my notes on the project at the time
June 26, 2025 at 11:32 AM
jessmart.in
iykyk
June 23, 2025 at 2:31 PM
jessmart.in
loved that piece, particularly these points.

as a result of reading:
- I'm trying to be aware of and document the adaptation I'm doing in order to get good results
- I'm working on creating better artifacts "aimed more closely at how real work is done"
The process of adaptation is often not obvious, even to the people doing it. There are lots of strategies and patterns and behaviors people pick up or develop tacitly as a part of trying to meet goals. Cognitive work that gets deeply ingrained sometimes just feels effortless, natural, and obvious. Unless you’re constantly interacting with newcomers, you forget what you take for granted—you just know what you know and get results. Tying this back to our LLM systems, what is imagined is powerful agents who replace engineers (at least junior ones), make everyone more productive, and that will be a total game changer. LLMs are artifacts. The scaffolding we put in place to control them are how we try to transform the artifacts into tools; the learning we do to get better at prompting and interacting with the LLMs is part of how they transform us. If what we have to do to be productive with LLMs is to add a lot of scaffolding and invest effort to gain important but poorly defined skills, we should be able to assume that what we’re sold and what we get are rather different things. That gap implies that better designed artifacts could have better affordances, and be more appropriate to the task at hand. They would be easier to turn into productive tools. A narrow gap means fewer adaptations are required, and a wider gap implies more of them are needed. The problem is that while the skills are real and important, I would argue that the level of sophistication they demand is an accidental outcome of poor interaction design. Better design, aimed more closely to how real work is done, could drastically reduce the amount of scaffolding and learning required (and the ease with which learning takes place).
June 19, 2025 at 5:05 PM
jessmart.in
I'm not a Fancy YouTuber™ 🤓, but I've done a lot of livestream coding on that site over the years (100+ hours!).

Finally got around to updating my "channel front page"

If you're into live-coding, you might like some of it.

www.youtube.com/jessamartin
June 19, 2025 at 2:02 PM
jessmart.in
Never a better time to build personal tools using agentic coding techniques. I'm working on a Trello replacement with integrated tool-using AI chat sidebar.

All of the features I want, none that I don't.
June 18, 2025 at 2:32 PM
jessmart.in
you all good!
June 14, 2025 at 9:09 PM
jessmart.in
Cloudflare gets it
June 12, 2025 at 3:58 PM
jessmart.in
I have seen the coming software apocalypse 👀

async SWE agents are real, ya'll
June 11, 2025 at 9:09 PM
jessmart.in
New Gemini Pro model checkpoint just want to cooooooodeeee

Me: please write a design doc for this feature in markdown file

Gemini:
June 6, 2025 at 4:25 PM
jessmart.in
this shot of me speaking at local-first conf is a caption contest begging to happen
June 6, 2025 at 3:19 PM
jessmart.in
still sad about "no Arc" but I know it's not your fault
May 30, 2025 at 9:13 PM
jessmart.in
Claude 4 found the error 😏
May 23, 2025 at 2:14 PM
jessmart.in
.base files basically act as a filter over your markdown files, where the properties are defined in the markdown file frontmatter.

The DSL looks pretty robust.

Very curious to see what people do with it.
May 22, 2025 at 12:42 AM
jessmart.in
In this example the base file is being visualized as a table AND a map. (I don’t think map is supported yet)
May 22, 2025 at 12:40 AM
jessmart.in
...and SimCity
May 12, 2025 at 6:32 PM
jessmart.in
Reading about Civilization I (1991) for UX inspiration.
May 12, 2025 at 6:04 PM