Ian Maurer
@imaurer.bsky.social
Fighting cancer with code.
Interests: Bioinformatics, Genomics, LLMs, Python, Rust, NLP.
https://imaurer.com/
https://github.com/imaurer/
Interests: Bioinformatics, Genomics, LLMs, Python, Rust, NLP.
https://imaurer.com/
https://github.com/imaurer/
I'm only here because @simonwillison.net posted about this on elon's plaything.
November 3, 2025 at 7:47 PM
I'm only here because @simonwillison.net posted about this on elon's plaything.
Built on top of the powerful open access API services of the NIH:
- ClinicalTrials.gov
- MyVariant.info
- PubMed / Pubtator3
www.ncbi.nlm.nih.gov/research/pub...
- ClinicalTrials.gov
- MyVariant.info
- PubMed / Pubtator3
www.ncbi.nlm.nih.gov/research/pub...
April 8, 2025 at 2:51 PM
Built on top of the powerful open access API services of the NIH:
- ClinicalTrials.gov
- MyVariant.info
- PubMed / Pubtator3
www.ncbi.nlm.nih.gov/research/pub...
- ClinicalTrials.gov
- MyVariant.info
- PubMed / Pubtator3
www.ncbi.nlm.nih.gov/research/pub...
Short video on YouTube:
www.youtube.com/watch?v=bKxO...
www.youtube.com/watch?v=bKxO...
Introduction to BioMCP
YouTube video by GenomOncology
www.youtube.com
April 8, 2025 at 2:51 PM
Short video on YouTube:
www.youtube.com/watch?v=bKxO...
www.youtube.com/watch?v=bKxO...
Github Repo (MIT License):
github.com/genomoncolog...
github.com/genomoncolog...
GitHub - genomoncology/biomcp: BioMCP: Biomedical Model Context Protocol
BioMCP: Biomedical Model Context Protocol. Contribute to genomoncology/biomcp development by creating an account on GitHub.
github.com
April 8, 2025 at 2:51 PM
Github Repo (MIT License):
github.com/genomoncolog...
github.com/genomoncolog...
Listening now... came here to say congrats. :)
January 14, 2025 at 9:11 PM
Listening now... came here to say congrats. :)
I did finish it and listened to your podcast with @mkennedy.codes
Good stuff. Subscribed on overcast.
Good stuff. Subscribed on overcast.
January 4, 2025 at 1:23 PM
I did finish it and listened to your podcast with @mkennedy.codes
Good stuff. Subscribed on overcast.
Good stuff. Subscribed on overcast.
Great article this morning. Got to the blog scroll link and… now I am here.
I hope I don’t get distracted too much longer. I want to finish it.
I hope I don’t get distracted too much longer. I want to finish it.
January 3, 2025 at 12:52 PM
Great article this morning. Got to the blog scroll link and… now I am here.
I hope I don’t get distracted too much longer. I want to finish it.
I hope I don’t get distracted too much longer. I want to finish it.
Your line colors are messing with my mind.
January 1, 2025 at 5:01 PM
Your line colors are messing with my mind.
You can’t for-loop taste
December 22, 2024 at 10:39 PM
You can’t for-loop taste
Reposted by Ian Maurer
Just everything else with AI, you can get good results with low effort.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
December 20, 2024 at 2:17 AM
Just everything else with AI, you can get good results with low effort.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
Just everything else with AI, you can get good results with low effort.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
December 20, 2024 at 2:17 AM
Just everything else with AI, you can get good results with low effort.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
Then you can grind and keep eeking out gains.
This has add’l benefits:
- improvement on that task
- greater vision of what else you can do
- higher expectations on what good results are
- better first drafts the next time.
How are they going to get the sun to shine like that in winter?
December 19, 2024 at 2:31 PM
How are they going to get the sun to shine like that in winter?
Perplexity did find me this quote:
“Upon inspection, we found that 100% of GPT 3.5 Turbo JSON-mode responses placed the "answer" key before the "reason" key, resulting in zero-shot direct answering instead of zero-shot chain-of-thought reasoning.”
ar5iv.org/html/2408.02...
“Upon inspection, we found that 100% of GPT 3.5 Turbo JSON-mode responses placed the "answer" key before the "reason" key, resulting in zero-shot direct answering instead of zero-shot chain-of-thought reasoning.”
ar5iv.org/html/2408.02...
Let Me Speak Freely? A Study on the Impact of Format Restrictions on Performance of Large Language Models
Structured generation, the process of producing content in standardized formats like JSON and XML, is widely utilized in real-world applications to extract key output information from large language m...
ar5iv.org
December 14, 2024 at 6:47 PM
Perplexity did find me this quote:
“Upon inspection, we found that 100% of GPT 3.5 Turbo JSON-mode responses placed the "answer" key before the "reason" key, resulting in zero-shot direct answering instead of zero-shot chain-of-thought reasoning.”
ar5iv.org/html/2408.02...
“Upon inspection, we found that 100% of GPT 3.5 Turbo JSON-mode responses placed the "answer" key before the "reason" key, resulting in zero-shot direct answering instead of zero-shot chain-of-thought reasoning.”
ar5iv.org/html/2408.02...
Ugh. Sorry to hear this.
December 11, 2024 at 5:27 PM
Ugh. Sorry to hear this.
That's my favorite fun fact about SQLite. Seems to be a competitive advantage for them securing contracts from large organizations.
Which, good on them, I think it's a more than fair way to do OSS.
Which, good on them, I think it's a more than fair way to do OSS.
December 10, 2024 at 7:48 PM
That's my favorite fun fact about SQLite. Seems to be a competitive advantage for them securing contracts from large organizations.
Which, good on them, I think it's a more than fair way to do OSS.
Which, good on them, I think it's a more than fair way to do OSS.
That’d be sweet. Thank Claude for me.
December 8, 2024 at 10:20 PM
That’d be sweet. Thank Claude for me.
Bunny CDN still treating you alright?
December 8, 2024 at 10:20 PM
Bunny CDN still treating you alright?
Nice looking script.
December 8, 2024 at 10:04 PM
Nice looking script.