JD Hyman (@jdhyman) header
JD Hyman (@jdhyman) profile picture

The Hyman Podcast • Accidental vibe coder 🤖 • ChatGPT Pro200 • Claude Max • X Super

Columbia, SC
Joined June 2015
Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)Photo by JD Hyman (@jdhyman)
Parroting fanaticism isn’t a good luck.
The leader of the free world on Truth Social today. He knows who has his back and who is here to subvert him. He knows. President Trump also knows how depraved these Woke Reich Islamic sympathizers are. It’s time to rip the bandaid off.
32
U.S. Congressional hearings are such a joke.
14
My usage is depleting and I’m not even using Codex.
1
22
About damn time.
GPT-6.2 Sol & Astra are *finally* here. We've also expanded ultrafast to Pro 200 plans for GPT-6.2 Sol, it's my go-to model for most day to day things.
20
Good god, @greggutfeld is mentally ill and has TV shows.
12
Careful @MeghanMcCain, your nepotism is showing...
Maybe he actually is the person to lead us out of this mess…
16
Sometimes change is necessary. America proved in this election they wanted change and that means change in congressional leadership too.
Democrats, we won the majority. But we still need to win the trust of the American people. We can’t return to the status quo. If you agree, join us at chrisforleader.com
13
What are people building and shipping, I want to know!
21
Claude and Grok are busy building and shipping while Codex and the Dot are sitting in the corner because if brains were gas it wouldn’t have enough to power an ant’s motorcycle around a penny one time.
33
This meme takes on a whole new meaning
24
This year I was HARDCORE OpenAI. Like I was DIEHARD! To the point where I went Pro200. Now--omg, I literally told my dot that it was fired because it slacked on basic task and all it could do is apologize. Almost all of my projects are in Claude and Cursor. I am still holding out hope through the end of my current billing cycle then I'm cutting the cord.
Your dot now has… ⚡️ Faster replies 💬 Less jargon 🫡 Better follow-through And, they can now follow your custom instructions. Tell your dot in the thread things like how to behave or which actions need your go-ahead. “always talk in lowercase.” “Show me a draft before sending any emails.” You can also set custom instructions and rules in Customize > Behavior on mobile, or through your dot’s menu on web and desktop.
1
1
55
Got tattoos? Free Chipotle!
we had to run it back w the ultimate chipotle bag @wizkhalifa 🏃🏾‍♂️ show ur tats & score a BOGO this friday the 13th valid 11/13/26, 3–4pm local time. in-restaurant only at participating restaurants in the U.S. & Canada. tattoo, as determined by restaurant, req’d. limit 5 free entrees/check. exclusions + additional terms: chipotle.com/tatted
24
I love Dolly but I'll never forgive U2 and Apple for that pre-loaded album without our consent.
U2 announces new single “Torn” featuring Dolly Parton out this Friday.
1
110
I officially have @claudeai duet! What should I build first?!
1
1
115
Anthropic is slowly but surely earning my business just upgraded to Max. My OpenAI subscription ends 11.17. Let’s see if I’ll be downgrading completely and moving my portfolio completely.
We wanted to do more for the people who’ve been with us the longest. So with Fable 5.2, the longer you’ve been subscribed, the more Fable resets you get.
1
38
Fable 5.2 dropped and in response, @OpenAI dropped features no one asked for and a handheld that sold out in 23 mins. Make it make sense.
You can now connect multiple local computers to the same dot and choose which one it uses for a task. It can build and test your app on a dedicated machine while you keep coding on your laptop.
21
Honestly still on my way out the door. Pro 200 is still basically unusable.
Saw some confusions, but all Pro plans have no 5-hour limits, and give the following usage compared to Plus. ✅ Pro 100 = 10X ✅ Pro 200 = 20X ✅ Pro 500 = 50X, access to ultrafast
10
Our open-source ML Intern is now available as a Slack bot you can install in your own workspace! Evaluate models, build datasets, and fine-tune models on the fly 🚀
YOUR data deserves YOUR own, private model. Now, let @HuggingFace's ML Intern build it for you. Connect your data, files, mcps. And call it from Slack.
93
44
19
341
45,814
Today we are rolling out Gemini Omni Flash 1.2, with: - longer video generation (15 seconds) - deeper cinematic controls for camera pov - support for multi-turn conversation editing - less visual drift - 30% lower price than Omni Flash 1.1 Read more: blog.google/innovation-and-a…
244
156
69
2,484
190,120
JD Hyman retweeted
Impressive paper from CMU. Soon, we will have coding agents that can tune the entire stack, from prompts to token serving. This work experiments with coding agents to tune an LLM serving stack. They get Claude Sonnet 5.5 to build a serving system for Gemma E4B that delivers 4.87x the value per GPU dollar of the vLLM reference. ServerBench asks coding agents to design a complete, specialized serving system and decide how many concurrent sessions to run. The score is latency-qualified token value per dollar of GPU cost, measured by replaying real coding sessions with their dependencies intact. An independent check verifies the system works. Across 11 coding-agent models and three deployments, 22 of 33 runs produced a verified system and 13 beat the strongest baseline. The best designs raised capacity and chose the load to match it. Failed runs often shipped optimizations whose new code paths were never tested. arxiv.org/abs/2611.06841 Chat with Paper: academy.dair.ai/papers/serve…
52
20
3
115
9,970