<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><description>Your friendly neighbourhood (data &amp; neuro-) scientist. Interested in healthcare, big data, social interaction, tea, and cats. 🇫🇮 living in 🇸🇪</description><link>https://bsky.app/profile/juulia.bsky.social</link><title>@juulia.bsky.social - Dr Juulia Suvilehto</title><item><link>https://bsky.app/profile/juulia.bsky.social/post/3mfmytltcks2x</link><description>Why do so many academics keep saying &#34;let&#39;s try to submit this next week&#34; when everyone can clearly see that the manuscript is not ready? I see many PhD students so stressed out about this when no one tells them &#34;next week&#34; deadline is aspirational, and not actually a real deadline.  #AcademicSky</description><pubDate>24 Feb 2026 20:26 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3mfmytltcks2x</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3m3rf5wy2ns25</link><description>Great talk by @hollycummins.com at Lindholmen developer day. As a data person and as someone who used to work with healthcare this resonated a lot: &#34;don&#39;t make human tell computer things the computer already knows&#34;. It is so key but so often neglected in practice (&amp;definitely a thief of joy) #databs</description><pubDate>22 Oct 2025 08:05 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3m3rf5wy2ns25</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3m2olcrkxes24</link><description>Thank you everyone for your suggestions! This is how I ended up plotting my LLM system output vs human annotation vs LLM-as-a-judge evals. Extra thanks to @libbyheeren.bsky.social for boosting my original question.  #databs #dataviz&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>08 Oct 2025 11:52 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3m2olcrkxes24</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lzy6u6ovdc2i</link><description>Any references for good visualizations of LLM evals? We have orig system output (yes/no), a human annotator&#39;s y/n and a number of different y/n evaluations. The aim is to see how good our original system is and how much of the issues do the diff LLM evals catch. How would you visualize this? #databs</description><pubDate>29 Sep 2025 14:10 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lzy6u6ovdc2i</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lqv4cgp6ys27</link><description>Academics, get thee some fuck off skills. Marketable skills that can help you fuck off from academia and get employed elsewhere. I see folx do multiple postdocs because they don&#39;t know what else they could do. Don&#39;t let that be you - make sure you have the ability to fuck off #academia #AcademicSky</description><pubDate>05 Jun 2025 20:00 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lqv4cgp6ys27</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3loxpdzw7fk2b</link><description>I seem to re-learn this truth at regular intervals: if you&#39;re looking for an explanation of how something slightly niche works, find a master&#39;s thesis related to the topic. Students (at least in the nordics) are expected to explain the whole thing and not leave steps out (unlike academic articles).</description><pubDate>12 May 2025 09:56 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3loxpdzw7fk2b</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lmdfhnm7g22w</link><description>I was studying at the uni and we had programming courses where you could get support from the TAs on the course&#39;s own IRC channel. I used Google reader to keep up with my favorite blogs. There was so little fluff or ads on websites, it was just raw html and original, stellar content 🥺&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>08 Apr 2025 21:15 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lmdfhnm7g22w</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lktbpbvtgs23</link><description>My first #tidytuesday #pydytuesday where I look at how different palmtree species are armoured. I challenged myself to work in plotnine and it&#39;s been a learning curve for sure! Code available here https://github.com/jtsuvile/tidytuesday_juulia/blob/main/palmtrees/palmtrees.py</description><pubDate>20 Mar 2025 18:00 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lktbpbvtgs23</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lkqbxtrssk2c</link><description>Only dating advice I can give is to always check for implausible dates (like 1900-01-01 or anything &gt;50 years into the future) as potential missing values in disguise. #dataBS #datasky&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>19 Mar 2025 13:27 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lkqbxtrssk2c</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3ljiagg4kj22z</link><description>I&#39;m trying to use a metric_set with accuracy together with cal_validate_logistic() from {probably}. I&#39;m having a hard time figuring out how to set the parameter &#39;truth&#39; (which typically is &#39;target&#39;). @topepo.bsky.social do you know if there is a vignette showing how to do this somewhere? 🙏 #dataBS</description><pubDate>03 Mar 2025 15:13 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3ljiagg4kj22z</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lj3a2pgbt22k</link><description>I&#39;m trying out a new thing in my collaboration with clinical researchers this time around: I created a Quarto document with a &#34;results section&#34; based on a dummy model and I&#39;m using it to get buy-in for the metrics &amp; plots before we actually fine-tune the model. Let&#39;s see if this works! #dataBS</description><pubDate>26 Feb 2025 11:02 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lj3a2pgbt22k</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lgphisiw7s2k</link><description>This is something I see a lot - people who know very little about AI want to use it to do every thing. People who understand AI better are much more reserved. I do my best to disillusion folks in my organisation..&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>27 Jan 2025 07:53 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lgphisiw7s2k</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lggl3atvrs2k</link><description>I no longer live in the country I grew up in. Every time I come back home I&#39;m floored by how easy dealing with folks (customer service, contractors) is when I a) get to speak my native language and b) actually understand the rules that the society operates by. Just 🤯</description><pubDate>23 Jan 2025 19:03 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lggl3atvrs2k</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lfv37eng6c23</link><description>&#34;the methodology seems to work well enough that you can&#39;t rule it out immediately, *even if might be garbage for your particular task due to the specific intersection of quirks of your project.*&#34; it&#39;s always the project specific quirks that get you.. #dataBS https://www.counting-stuff.com/labeling-things-by-hand-when-everyones-trying-not-to/</description><pubDate>16 Jan 2025 20:03 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lfv37eng6c23</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lejvgj7lns2m</link><description>I gave a talk earlier in the fall about AI in healthcare, talking about the hype and the challenges and where does that leave us. My talk focuses on a Nordic single-payer style healthcare but a lot of it generalises to different healthcare systems https://youtu.be/qqcYHWlCKkk?feature=shared #healthcare  #dataBS</description><pubDate>30 Dec 2024 15:55 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lejvgj7lns2m</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3ld75gx5ra22n</link><description>My colleagues are impressed by my shitposting game on the designated shitposting teams channel when really I just repost the best of #dataBS&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>13 Dec 2024 15:54 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3ld75gx5ra22n</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lcctcvslns23</link><description>...and just like in Harlow&#39;s monkey experiments, IRL interaction with conspecifics is much, much better than either of these&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>02 Dec 2024 09:39 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lcctcvslns23</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lbsmchpomc2f</link><description>Playing around with the new Mall package mlverse.github.io/mall/ by @theotheredgar.bsky.social. Such a cool concept! But I cannot seem to figure out how to make my own data frame (CSV imported via polars) into the correct data type (which I suspect is MallFrame). Any tips? #dataBS&#xA;https://mlverse.github.io/mall/</description><pubDate>25 Nov 2024 22:50 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lbsmchpomc2f</guid></item><item><link>https://bsky.app/profile/juulia.bsky.social/post/3lbhfmpsjd226</link><description>Loving the @nhsrcommunity.bsky.social #RPySOC conference and would warmly recommend it to all of us health/hospital #dataBS people. It&#39;s my second year following it online and the useful info / unit of time ratio is the highest I&#39;ve found in this space https://nhsrcommunity.com/conference24.html</description><pubDate>21 Nov 2024 11:52 +0000</pubDate><guid isPermaLink="false">at://did:plc:o7pwq77io66e34l5vrs6mdue/app.bsky.feed.post/3lbhfmpsjd226</guid></item></channel></rss>