Ben Lesh ✓
@BenLesh · Austin, TX · joined 11 Mar 2009
Formerly Netflix, Google, and fin tech. Currently working in the AI space. Created plugins and MCP apps examples for OpenAI. RxJS project lead.
62 811Followers
1 580Following
52 524Posts total
107.4KViews on collected posts
Ultimi post
Cerchiamo nuovi post…
I'm really starting to wonder if coding harness maintainers have added system prompts like:
"Be as verbose as possible and think of as many ways as possible to keep looping on things to use tokens and make us money"
😂
10.6K views · 334 likes
· 6 reposts · 35 replies
Open on X →
I think some of the respondents are missing what I’m saying.
I’m not saying “no frameworks or libraries”
I’m saying “new frameworks and libraries designed with models as the targeted user are going to win”
4.5K views · 21 likes
· 2 reposts · 6 replies
Open on X →
@BenLesh @matthewcp Yeah I’m betting on that pretty hard.
https://t.co/CS32L01WQn
487 views · 26 likes
· 0 reposts · 1 replies
Open on X →
@BenLesh I’m not sure if I agree. Frameworks and libs take care of a lot of plumbing work behind the scenes. Why would I want vanilla JavaScript to reinvent that.
1.4K views · 11 likes
· 0 reposts · 3 replies
Open on X →
@steveklabnik I’m at the Austin Systems meetup 100% by chance and they mentioned you. 👋 I haven’t seen you since Austria with @bmeurer ❤️
2K views · 3 likes
· 0 reposts · 1 replies
Open on X →
@BenLesh Ahh - you're saying frameworks with clearer api's and type safety will survive better?
1.9K views · 12 likes
· 0 reposts · 2 replies
Open on X →
I am here to predict the (partial) death of your favorite JavaScript frameworks and libraries…
In the age of agentic coding, I seriously doubt that frameworks that provide a lot of implicit magic that helped humans be productive will be the successful frameworks. IMO, code
33.7K views · 182 likes
· 8 reposts · 44 replies
Open on X →
My OpenAI contract has wrapped up, and I'm looking for my next role.
For nearly two years, I've been building robust AI chats, MCP tools, AI UI integrations (MCP Apps), and more. I wrote some of the first Codex plugins: Their plugin evaluator, a comprehensive data visualization
16.9K views · 47 likes
· 5 reposts · 6 replies
Open on X →
@BenLesh @ThisDotLabs @coreyching Love seeing AI move from make a chart, to actually helping shape the thinking behind complex visuals!!
Crazy good work 👏
233 views · 2 likes
· 0 reposts · 0 replies
Open on X →
12/ It’s been fun working with @OpenAI at @ThisDotLabs.
This project sits right in my lane: AI, TypeScript, and data visualization. It’s been a blast turning that overlap into something OpenAI and Codex can actually use.
701 views · 0 likes
· 0 reposts · 1 replies
Open on X →
11/ Final clip: A geospatial visualization with a lot of features cooked up in a few hours!
The thing I’m hyped about: Codex can now help think visually before it builds visually. https://t.co/bfCRs4NGjA
749 views · 0 likes
· 0 reposts · 1 replies
Open on X →
0:55
10/ The architecture explorer handles topology, schemas, and state machines with zoom/pan, routing, filters, criticality, and readable layout rules.
Useful when “diagram this system” needs more than a quick Mermaid sketch. https://t.co/xkcVjsG9jp
321 views · 2 likes
· 0 reposts · 2 replies
Open on X →
0:47
8/ Under the hood, the plugin has specialist guidance for D3/SVG, Canvas2D, WebGL/Three.js, geospatial maps, dashboards, Gantt schedules, UML/software architecture, uncertainty, accessibility, testing, reports, PDF generation, slides and more.
353 views · 0 likes
· 0 reposts · 1 replies
Open on X →
9/ The Gantt work covers the real project-management mess: milestones, baselines, dependencies, critical path, WBS, resource load, import/export thinking, and editable timelines. https://t.co/b1ZjdcsR84
325 views · 0 likes
· 0 reposts · 1 replies
Open on X →
0:40
7/ The plugin also handles serious editorial work.
The Ukraine infographic forced source/method ledgers, dated map states, separate evidence categories, humane framing, and a clear core claim: “The front moved less than the war changed.” https://t.co/12ed8kRSxU
390 views · 0 likes
· 0 reposts · 1 replies
Open on X →
0:22
4/ The earthquake demo is the flashy one: a Three.js globe, USGS events, magnitude/depth encoding, selected-event focus, live source rail, reduced motion, and a geological cutaway that explains epicenter, hypocenter, and depth. https://t.co/H3Ma9FKRP3
559 views · 1 likes
· 0 reposts · 1 replies
Open on X →
1:00
5/ The soccer demo is a synthetic Chicago Lakefront FC scoring-influence network.
It uses the pitch as the analytical surface, animated pass chains, player ranking, shareable URL selections, and generated fictional team/player assets. https://t.co/gbpXUFx2XA
414 views · 1 likes
· 0 reposts · 1 replies
Open on X →
0:36
2/ The big idea: Codex should not jump straight from prompt to chart.
For complex visualizations, it can generate a visual design concept first, show it for approval, then implement against that concept as a semantic contract. In this thread, I’ll show you the generated https://
2.1K views · 2 likes
· 0 reposts · 1 replies
Open on X →
0:28
3/ That matters because data viz is not “pick a chart library.”
It’s analytical intent, evidence hierarchy, source/caveat handling, interaction model, renderer choice, accessibility, and QA. The plugin gives Codex a workflow for all of that.
1.1K views · 1 likes
· 0 reposts · 1 replies
Open on X →
1/ At @ThisDotLabs, I built a Codex Plugin, called Build Web Data Visualization with DevEx Lead @coreyching from OpenAI.
With this plugin, Codex uses image gen to design the app and then implements it really well. See the examples below. Claude can’t do this!🧵 https://t.co/GN6t
26.3K views · 53 likes
· 12 reposts · 4 replies
Open on X →
Rispetto ad account della stessa dimensione
9 post degli ultimi 90 giorni, accanto alla fascia di 10K–100K follower. copertura nella norma per la sua dimensione, reazione più debole della media.
Visualizzazioni mediane2 464questo account914mediana per 10K–100K
Copertura, %3.92%questo account3.38%mediana per 10K–100K
Interazione, %0.73%questo account1.53%mediana per 10K–100K
| Metrica | Questo account | Mediana per 10K–100K | Rapporto |
|---|---|---|---|
| Visualizzazioni mediane per post | 2 464 | 914 | 2.70× |
| Copertura (visualizzazioni ÷ follower) | 3.92% | 3.38% | 1.16× |
| Tasso di interazione | 0.73% | 1.53% | 0.48× |
Altri account di questa fascia → Confronta con un altro account → Come sono costruiti questi parametri →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
35326 May
321
749
701
233
16.9K8 Jul
33.7K16 Sep
1.9K
2K
1.4K17 Sep
487
4.5K
10.6K18 Sep
2.5K
Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.28%26 May
1.25%
0.13%
0.14%
0.86%
0.34%8 Jul
0.72%16 Sep
0.73%
0.20%
1.00%17 Sep
5.54%
0.65%
3.58%18 Sep
1.01%
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.53%.
What the audience does
Likes71.5%720 in total
Reposts3.3%33 in total
Replies11.5%116 in total
Quotes1.5%15 in total
Bookmarks12.2%123 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
Account simili
🇿🇦Zwidenya ✓
@zwidenyapT
125.6K followers
· 423.2K posts
Dog Lovers ✓
@DogLovers03
125.6K followers
· 6.9K posts
Ywuria 👁️🗨️🪝
@ywuria
125.6K followers
· 32.6K posts
Texas UIL ✓
@uiltexas
125.6K followers
· 16.5K posts
OlympusDAO |Smart, Money|
@OlympusDAO
125.6K followers
· 3.4K posts
Surya ✓
@goldspecter
125.6K followers
· 10.9K posts
Students For Liberty ✓
@sfliberty
125.6K followers
· 41.2K posts
Oscar Defi ✓
@OscarDefii
125.6K followers
· 350.2K posts