How People Are Using AI For Visualizing data

I am finding ChatGPT agents to be useful. They are a better fit with the "intern" analogy than any former AI - requiring oversight, still saving lots of time overall. For example, I update an AI cost/performance chart frequently. The agent did all the grunt work, with guidance. https://t.co/AGs7DRNxSh

ChatGPT agent: "create a PDF of a novel D&D adventure, add illustrations, make it super interesting and deep, add tables, etc" "Fix the formatting, build it out more" Got a 19 page PDF. Agent doesn't do layouts well, but pulls off building a coherent adventure, hard for LLMs. https://t.co/OYEo3L1hgP

My non-programming linguist wife is now a proud user of Ghostty, wherein she commands Claude Code to help her create charts and statistical analysis for her research. What a time to be alive. https://t.co/c9jQO06DUN

Interesting way I've been using Claude + journaling for those curious: Back in January I gave Claude 200,000 tokens, it was my last 6-months of journal entires. From there, I told it to predict my future. Given my previous 6-months, I told it to plot my best timeline + my darkest timeline along with the events that'd lead me there (like a movie). For each plot point, it also ranked my happiness out of 10. The dark timeline was extremely scary. (I had just got back to SF after a hiatus, and it talked about how my old-self would resurface, comparison traps, etc). The positive timeline was of course, preferred. What's really cool is when I did this back in January I became 1000X more aware of how I could potentially dig myself into a mental hole over the following 6-mo. Just by being aware of it, I was able to kinda steer myself way away from that bad timeline/mental traps. Kinda spooked me into being more aware of my flaws. In life its hard to understand how our mental states + events today (ex. collapse of a company, a breakup, a new job, etc) will chain together and affect us down the line. Kinda beautiful how these models can help us literally see/visualize potential futures and learn from them. Will def be doing this every 6-months for myself!! P.S: 6-mo later -- I def feel like I am living the best timeline it predicted :). shoutout claude. P.S.S: someone should build this into the best journaling app in the world. power of ai isnt "chat with my diary" I think its stuff like this.

Karpathy today said Cursor for Slides needs to exist.. but it already does. When asked to "Create a detailed data-driven slide deck based on Google's recent financial filings", It created a stunning 6 page deck with graphs, diagrams IN Google theme! It's called Genspark. https://t.co/Wy0DKVvFzK

Claude-4 is crushing data analysis and data enrichment for my team. Example use case: my team is running a webinar. Within a few hours of launch, we had thousands of signups. I grabbed only the timestamps of signups. Threw them into ChatGPT and Claude asking for a graph of signups per hour, benchmarks, insights, action plan. ChatGPT (specifically o4-mini-high) created errors and blamed them on me. Claude 4 handled it beautifully, first attempt. One of the biggest improvements in these models is going from 3-4 asks in conversation to 1 ask. And when things can be nailed in the first attempt, fully autonomous systems start looking more and more likely in 2026.

AI is totally getting out of hand We were internally testing things and it made this PhD level research-paper style graphs with one prompt 🤯 https://t.co/1MsuSA532Q

"o3, look up these books and give me a table of Amazon reviews, a one-sentence witty summary of the plot (no spoilers) & a link to where to buy them from an indie bookstore" Yes, the links all work. But impressive stuff you might miss is that it figured out half-obscured covers. https://t.co/9cLe6XdFkW

Deep Research with Gemini 2.5 has become very good. It spontaneously generates tables, scenarios, and compiles evidence. Haven’t spotted errors in spot checks. https://t.co/4JjVulx8v0

ugh time to add google analyt... oh wait... prompt to visitor analytics dashboard (one-shot with replit agent) https://t.co/DTQ4Lsg7Lg