Christos Argyropoulos MD, PhD
ChristosArgyrop@mstdn.science
<p>Division Chief, Nephrology, University of New Mexico<br />Posts: <br />Data Science, Bioinformatics, Computing, Statistics</p><p><a href="https://mstdn.science/tags/RNAsequencing" class="mention hashtag" rel="tag">#<span>RNAsequencing</span></a> <a href="https://mstdn.science/tags/Nanopore" class="mention hashtag" rel="tag">#<span>Nanopore</span></a><br /><a href="https://mstdn.science/tags/GLMMs" class="mention hashtag" rel="tag">#<span>GLMMs</span></a> <a href="https://mstdn.science/tags/Rstats" class="mention hashtag" rel="tag">#<span>Rstats</span></a> <a href="https://mstdn.science/tags/Bayesian" class="mention hashtag" rel="tag">#<span>Bayesian</span></a><br /><a href="https://mstdn.science/tags/Perl" class="mention hashtag" rel="tag">#<span>Perl</span></a> <a href="https://mstdn.s
Posts
-
View post
I regret asking the clunker (GPT5.6 Terra) to help with syncing the README.md file of my repo with the existing codebase. It now speaks like a lawyer or a drug commercial
-
View post
I find the #AI clunkers useful when automating performance benchmarks using 3 context markdown files: โ๏ธ#JSON related one about configuration/build/run matrix ๐โโ๏ธ#Perl to run the benchmarks/collect telemetry ๐#Rstats to visualize/analyze Even cheap/free models work!
-
View post
Very pleased with Kimi K.3 which I took for a ride today via #GithubCopilot. Good questions, solid plans, efficient, token frugal implementations in a multilanguage project with multiple branches: C, #Python, #Perl, #rstats #make. It handled documentation needs too. #KimiK3 FTW
-
View post
So now that #AMD removed support for gfx906 in ROCm, we could hope to buy some kick ass cards at reasonable prices in the secondary marker ?
-
View post
Hahaha during refactoring the clunker eliminated an underutilized codepath but important ; we are living the dream ๐คฃ๐คฃ๐คฃ
-
View post
Time to regulate GEMM and GEMV for safety in the name of #AI alignment. Anyone using #BLAS without a license especially if they are running models locally will go to Gitmo
-
View post
This is going to be a long night, the chatbots need to be reminded frequently that #Rstats + #Perl &gt; #Python #Bioinformatics #AI
-
View post
There is wisdom in buying 2 Hirsch/yubi keys at the same time. Having FA the purchase part, I have just FO the wisdom
-
View post
The AI economics math are not mathing (longer version of a short tweet/post/toot) https://christosargyropoulos.substack.com/p/the-ai-math-are-not-mathing?utm_source=share&amp;utm_medium=android&amp;r=1tfbmy
-
View post
Kimi K3 &quot;bribing the hamster&quot; hits again! Praise the #AI lord, AGI is here
-
View post
Reddit is the mother that keeps giving #AI
-
View post
Life sciences has been hallucinating facts and theories for a while and it is only through experimentation we learn. AI will just connect plausible, probable, factually false and truth statements into a grand hallucination. Live your Myth in Fable
-
View post
The #reproducibility of p-valies is terrible (we derived the distribution of p values for GLMs with Andy Grieve through a Bayesian interpretation (ฯ letter) here: https://arxiv.org/abs/2305.00636). I don&#39;t understand why people keep noise mining p&#39;s #stats Formula &amp; graph ๐
-
View post
#Perl as a Data Science Language https://www.iconoclasts.blog/chrisarg/perlasadatascilang.html.en