Co-Founder @buildpurdue (Founder/Builder Community) + Founder Polyidus + Student @LifeAtPurdue
Joined July 2025
- Tweets359
- Following122
- Followers38
- Likes295
Ok, why do zero-sugar sodas have calories in them? Like, what do you mean you're going to put 5 calories in this... Also, since when is this a thing, like why did zero sugar start meaning zero calories.
Choose one please!
U think REA.tools will work on the human body?
this might be my last normal tweet before biotech goes vertical
i'm not a whistleblower. I don't work at Lilly, Novo or any of the big labs
but there is a massive shift happening in human biology right now.
it's sitting in press releases and almost nobody reads them
the last 6 months alone:
- a once-daily pill roughly doubled median survival in metastatic pancreatic cancer (13.2 vs 6.6 months)
- a one-time gene editing infusion cut hereditary angioedema attacks 87%, and 62% of patients stayed attack-free and off all other prevention for 6 months
- a one-time gene edit for cholesterol dropped LDL up to 62%, and it held for over a year
- a personalized mrna vaccine beat keytruda alone in a 1,137 patient melanoma phase 3
- retatrutide hit ~28% weight loss in a phase 3
- an ai-designed drug entered phase 3
- a pill that regrows hair passed its second phase 3
- an immunotherapy that could let some rectal cancer patients skip surgery, chemo and radiation just got fda approved
- a man lived 271 days off dialysis on a pig kidney
- human cells were caught passing chromosome-sized chunks of DNA to their neighbors, and it changes how the neighbors behave
we are on the verge of an era where the human body is a software problem and the update cycle gets shorter everyday
stick around, it's going to get weird
bio/acc
Elliot Drel retweeted
Day 25/
We are doubling all usage limits for all Pro 100/200/500 tiers and Business Premium, effective today.
We heard your kind feedback and won’t be reducing usage limits for Pro 200, so your Work and Codex usage will stay the same. We’re also going to continue offering extended limits for dots for now, and usage of your dot will still continue to be separate from your default usage.
And we'll of course also keep shipping better models and other efficiency improvements that increase your effective usage over time.
SLI5 of The Midwest College Rankings 2026 by @startmidwest, using data from @dealroomco (article in comments):
Which Midwest universities produce the most successful startup founders?
Here are the biggest takeaways:
🏆 The top 3 remain unchanged: University of Chicago (#1), Northwestern (#2), and University of Michigan (#3) lead the region.
🚂 Purdue climbs to #9: @LifeAtPurdue moved up one position, making it the second-highest-ranked university in Indiana, behind Notre Dame (#6). Great to see the continued progress of Purdue's entrepreneurial ecosystem due to the work of @PurdueInnovates, @PurdueHIVE, @buildpurdue, The Anvil, Kostas Grigoriou, and many more.
📊 Founder density matters: Northwestern leads with 13.26 founders per 1,000 alumni, followed by UChicago at 12.68. The rankings account for university size rather than simply rewarding schools with the most graduates.
💰 Funding is a measure of success: More than 17% of startups founded by UChicago alumni raised over $10M, compared with 16% for Michigan and Notre Dame.
🌎 Midwest talent has global reach: Michigan is the only university system from the six-state region among the world's top 20 for total alumni founders. Many successful founders educated in the Midwest build their companies elsewhere.
My biggest takeaway:
The Midwest is clearly capable of producing successful founders. The question is how we get more of them to build their companies here.
I think a big part of the answer is making it easier for founders to find each other, access resources, and build strong communities.
That's something I've been thinking about a lot through my work at Purdue. There's so much entrepreneurial talent here, and I think we have a huge opportunity to better connect the people who want to build something.
T3 code just figured out AI thread co-working. Game over
New beta feature in T3 Code: "share with guest"
I'm so tired of copy pasting from a thread, sending it to a coworker, waiting for them to respond, screenshotting the context and sending it back to my agent.
Now I just send them a link.
It's a silly place to start, but I was heavily inspired by the Uber Eats/Doordash group order flow. If I send someone a link to my group order, they can add whatever they want, but ultimately I decide if the order goes through or not.
Realized that flow would solve at least half the use cases I have for "shared threads", so it's now built into T3 Connect in T3 Code (nightly)
Very curious how you guys feel about this flow. It's still early and I expect to adjust it a bunch (I left the ui bad so @maria_rcks will come in to fix it sooner)
Also s/o Opus 5.5 for making this lil video about the feature :)
Elliot Drel retweeted
T3 Code version 0.1.3 is now live! Stable is fully caught up to the nightly :)
Pretty small release, here's what's changed since v0.1.2:
- Complete redesign of the app to look significantly better
- Share threads with a guest in 2 clicks
- Opening a thread loads 85% less data, and cold connects send up to 93% less.
- Cmd+K search is comically better.
- T3 Connect promotes to LAN for using other boxes on same network
- File preview overhaul, now supporting office files, mermaid diagrams, sqlite dbs, audio files, pdf annotations, svg attachments, and more
- T3's agent instructions are 60% shorter, and the MCP tool schemas are a third smaller.
- "steer" is default (queueing sucks)
- More reliable PR linking
- You can now add custom instructions for title gens
- Muse Code now shows in usage tab
- All emdashes removed forever
- Various accessibility wins for screen reader support
- Code highlighting got way faster with @.pierre/highlights
- PR timeline, code tab, summary and search got way faster
- Streaming text updates are batched, so long runs use less CPU
- ~480 other things got merged that I forgot lol
...okay maybe it's a big release.
Best use case of @ChatGPT for college students. Give ChatGPT (or Codex) your graded exams and ask it to check for grading mistakes and if you should appeal anything.
Prompt: Review this exam and flag any mistakes in grading or inconsistencies that I should appeal.
Chat just found me 10 points on my accounting exam and 15 on my operations anagement exam.
Eveyone struggling with ai not outputting shippable features. Read this article by @poteto about verification environments.
Your agents need the capability to interact with the product the same way you users will. Otherwise how can you except them to create working features.
Imagine if you couldn’t ever try the feature, but just had to read the code and pray. Oh and belittled when it’s doesn’t work.
Tell me the last time Claude did the full job. Codex Sol 6.1 does the FULL job, every time. , even when you asked.
In this thread, it is teaching me content for an exam I have, while reading the class content for the next topic to teach me. It figured out that the teacher posted a practice exam, so it took the time to read that too and updated the learning plan to be more in line with what will actually be tested. Then, while it was doing that, it also noticed that the assignment due date moved from last night to today, which it also updated that in my task manager system.
Notice, I didn’t tell it to do anything, it decided to do this ON ITS OWN. I didn’t get 10 questions asking for approval, I didn’t get an essay after explaining why it didn’t want to do the work.
It just did all the work, thoroughly double-checked to make sure it didn’t miss anything, and then told me the final outcome.
And for everyone who says "opus 5.5 is better", yes it may write slightly, marginally better code, sooooo just have Codex use Claude to write code. Then you will have the best of both worlds. The SOTA model to write the code and the end-to-end task completion of Codex.
Btw for everyone wondering, the reason Codex is so good at doing the task end to end is because that’s just how their model is built, it is a fundamental difference of the mentalities behind how the models work that leads to outcomes like this. In my opinion, this is a superior mentality.
Switch to @openai Codex!
Codex Composer is on another level. For anyone who was hating on Codex Composer, saying that Claude Code has had this for months, you need to try the ChatGPT desktop app. Your life will change!
The composer somehow remembers the rules and guardrails that I gave it throughout the conversation (and I have 30+ turn sessions).
It is terrifying to me how often it says word for word what I was about to type.
