@david_laprade

Husband, dad, engineer, biologist, runner, ex-philosopher. Shouting into the void at https://xitter.cf/t.co/2adtZo9Ooa. Eng at @StandardModelHQ. Opinions my own

laprade.blog
Joined October 2014
  • Tweets196
  • Following109
  • Followers105
37Media
He's going to have another kid before he's 50, 95% confidence
44
tell me you distilled claude without telling me you distilled claude
40
Claude is consistently worse for me than all the other leading models. For code, for research, for everyday things. Has been for over a year. I genuinely don't know why everyone likes it so much
1
67
Example from just now. I gave the same prompt to gpt 5.1, gemini 3 pro, and opus 4.1 thinking. gpt and gemini actually did what I asked (gpt was better), claude instead wrote me ... some documentation
1
92
another example, AWS change needed, same prompt to codex 5.3, opus 4.6, and gemini 3 pro. Correct answer was "use a step function", codex and gemini got it, opus did not
41
In June @SendGrid sunset my legacy plan and converted it to a 60d free trial without telling me. 30d after expiration, without warning, they deleted all of my data. Every email that has subscribed to my blog over the last 10 years, gone. Incredible
60
not to give it away but...
Standard Model Bio@StandardModelHQ17 Dec 2025
On our Substack today: we tested our Standard Model on data from @MSKCancerCenter during the iHub Challenge 2025 Cohort Program—and we’re releasing the model weights for public use. Blog post: blog.standardmodel.bio/p/the… Docs & quickstart guide: docs.standardmodel.bio/

The Patient is Not a Document: Moving from LLMs to a World Model for Oncology (Part 2)

We Tested Our Model with Memorial Sloan Kettering Cancer Center’s (MSK’s) Data while Participating in the MSK iHub Challenge 2025 Cohort Program, and We’re Releasing the Model Weights to the Public

blog.standardmodel.bio
1
35
Soooo excited to be a part of this!!
Standard Model Bio@StandardModelHQ30 Sep 2025
We’ve been pretty quiet on here so far, but today we’re excited to introduce ourselves and share three new papers on foundation models in bio. 👋 Come take a look at our Substack post to learn more, and follow us here to receive future updates, too. 📥 bit.ly/46tdvRk
1
1
35
Can't we all just go back to making fun of the Coldplay couple? Life was better then
58
We recently got 5+ quotes for some work on our house. All were in the same ballpark except one, which was outrageously high by >50% for the same work. Completely out of touch with the market. Yet this company is clearly doing great. How?
1
43
more replies
Back to the construction company. It works because their process weeds out all of the bad people right away. Like us. We are objectively bad construction clients. We're going to be demanding. We know what stuff costs, we're going to negotiate, we care about details
1
25
They don't want to work with us. They want people who don't know prices, who aren't going to litigate everything, who just want a smiling face to give them an okay result. The very top of their funnel guarantees that this is who they work with. It's a good strategy!
23
It's sadly already gotten to the point where I assume most clips I'm seeing are AI generated. It's really killed animal videos for me
Nemo
@thecaptain_nemo19 Aug 2025
realized this was entirely AI created after searching for the source for 30 minutes things are going to get weirder and weirder
1
38
discovered a mind-blowing trick this weekend that's a total game changer. It completely breaks the internet. Here are four wild examples 🧵
2
56
Go to settings > mute and block > muted words Create entries for: - mind-blowing - game changer - broke the internet - wild example Now your feed will no longer be filled with idiotic AI click-bait 🙂
1
1
51
the difference between consistency and insanity is subtle
34
psychological principle that I've been appreciating lately: the same action performed with different intentions can have very different psychological results
1
47
e.g. if you do something for a reward vs for its own sake you're less likely to want to continue doing it
1
39
in insomnia literature people talk about "sleep efforts", doing things with the intention of sleeping better, and *the intention alone* is enough to perpetuate sleep disorders
27
Load more