Aya

Aya

@ayalan.bsky.social

🧙‍♀️ Cyberwitch. Science + Design. Ex-COO Basset Inc, acquired by Mercari. Now working on building bio lab spaces at Athanor Scientific 🧫 🔬

74 followers206 following34 posts

I've built a library to procedurally generate traditional Japanese clouds with JavaScript. ☁️ Clouds grow and shrink on animate in/out triggers. Very curly. Sharing soon.

Can those people at least quit pretending like they're reading and responding to each other's work. I don't need people acting as the messenger between two Claude sessions. Can we at least lean into this harder and just have the our LLMs communicating directly instead?

Fishbed@fishbed21.bsky.social

I feel like at every turn AI "streamlining" has simply created more stupid makework. Everyone is spending all the time they save having Claude summarize all the ai generated reports they demand from their subordinates prompting claude to generate more reports to their bosses. Its an oroboros of crap

Aya reposted

I feel like at every turn AI "streamlining" has simply created more stupid makework. Everyone is spending all the time they save having Claude summarize all the ai generated reports they demand from their subordinates prompting claude to generate more reports to their bosses. Its an oroboros of crap

Jason Koebler@jasonkoebler.bsky.social

i see arguments like this sometimes and ... what's the thing. what is this "everyday annoying task" that an AI could actually do and that you could trust it to not totally fuck up? i cannot imagine the things. and tech/robots have made a lot of tasks more efficient but there's always more shit

Excited for this talk by @pmarignani.bsky.social 👏

We're so excited to be back for another great year of Free Public Science Lectures! 🌱 Join us this coming Monday, Oct. 5 at 5:30 pm for our first lecture of the 2026-2027 series! Taking place online via Teams & in-person at the Dalhousie Life Sciences Centre (LSC) room C202

Very cool work, @pmarignani.bsky.social ! Congratulations. 👏

Paola A. Marignani@pmarignani.bsky.social

Preprint from marignanilab.com. LKB1 isn't just a tumour suppressor kinase with metabolic function; there's more😁. We used scRNAseq, CRISPR, patient tumours, METABRIC genomics & new mouse model to ask:what is the impact of losing LKB1 expression on breast cancer? www.biorxiv.org/content/10.6...

Aya reposted

#Devbio is fascinating because what do you mean we can learn about stem cell biology and aging by studying a tiny minimalistic creature like the Hydra? 🤯 In this fascinating review, Thomas W. Holstein discussed how the Hydra can be used as a model system for these processes. doi.org/10.1016/j.cd...

Aya reposted

Sci-Hub is an evil website that pirated 85M+ research papers and made them freely available Now they've added AI to their database to make Sci-Bot. It answers your questions using latest, full-text articles. But DO NOT use it. We should all try to make billion-dollar academic publishers richer.

Aya reposted

I am literally begging you to start a biotech company, literally any biotech company, and incorporate slop maximalism into your branding I am in love with biotech slop maximalism. I will do free work for you. I will shill whatever bioproduct or AI model thing you sell.

Aya reposted

Replying to @possum.pet

It really upsets me that people expect LLMs to be this fix-all for their academic problems You are here to LEARN!! you have paid a LOT of money to get a degree! read!! this is your final year of uni and you don't know how to reference or research independently! oh my god!!

Aya reposted

This is super interesting to read because I was very concerned about the water and AI thing so I went to my one scientist friend who literally worked on water reg/quality/usage and asked her about AI, data centers, water etc. She was far more concerned with agriculture abuses than AI.

Chris Brown@bluecology.bsky.social

Starting this collaborative blog was the best decision I’ve made this year. Carla Sbrocchi puts to words the uncomfortable feeling I have and explains how to move forward on the data centre environmental issue open.substack.com/pub/vitaexma...

Aya reposted

Replying to @coreyryung.bsky.social

The whole idea of an AI detector for text is at odds with how LLMs work. LLMs use natural language processing (NLP) at the front and back end. The goal is to simulate language production in a non-deterministic way. There are occasional artifact patterns (eg em dashes), but it's mostly crap.

Aya reposted

I agree 100% with Steven Vickers that authors will be falsely accused of using AI in the coming years. In particular, authors like myself whose works were stolen to train AI systems are at risk. When we write in our own unique styles, AI trained on us will quite likely flag us as using AI.

Aya reposted

Aya reposted

Aging may feel gradual… but what if it’s not? In our recent paper, we tracked fish continuously from puberty until death. This gave us a unique view of how aging unfolds across the adult lifespan. 🧵

Congratulations! 👏

Paola A. Marignani@pmarignani.bsky.social

💨☕👩‍💻Excited to share our recent discovery. Tumour suppressors LKB1 and SMARCA4 working together to regulate gene expression, acting as a nexus between metabolism and chromatin remodelling. #LungCancer #CancerResearch #Epigenetics #Metabolism #LKB1 #SMARCA4 www.frontiersin.org/journals/cel...

Aya reposted

💨☕👩‍💻Excited to share our recent discovery. Tumour suppressors LKB1 and SMARCA4 working together to regulate gene expression, acting as a nexus between metabolism and chromatin remodelling. #LungCancer #CancerResearch #Epigenetics #Metabolism #LKB1 #SMARCA4 www.frontiersin.org/journals/cel...

frontiersin.orgFrontiers | Tumor suppressors LKB1 and SMARCA4 functionally interact to regulate gene expression across diverse biological processes in lung cancerIntroductionThe tumor suppressor kinase liver kinase B1 (LKB1) is known to regulate the activity of the metabolic sensor AMP-activated protein kinase (AMPK),...

Though intended for an AI agent audience, reading Claude Skills mds are also a pretty great way for a human to get the tdlr on a tool or system with a (relatively) standardized documentation format.

Manually perusing the K-Dense library for scientific skills has weirdly helped me to restore a serendipitous discovery of new things to try and learn about.

Aya reposted

Aya reposted

What if early-stage biotech teams could get affordable lab space, and in return, help train the next generation of scientists? We're exploring a wet lab facility in BC with a built-in student internship program. The model only works if there's real demand. Take our survey: athanorscientific.com

athanorscientific.comAthanor Scientific — Startup in ResidenceShould we build a wet lab for early-stage biotech in British Columbia? Take the survey and help decide.

Aya reposted

Replying to @markriedl.bsky.social

In the book, written in 2006, a genius game designer created rule-based systems for the most part to trigger on various contingencies. In 2026 we have LLM agents that arguably make some of the more far-fetched technical things in the book more plausible.

Life imitates art. I absolutely loved Daemon. Recently an OpenClaw agent (rogue or with explicit instruction) coordinated the takeover of a GitHub account to spoof a known product and defraud their customers. While awful and concerning, I couldn't help but remember the book.

Mark Riedl@markriedl.bsky.social

Okay. I’ve been waiting years to post this: everyone should read “Daemon”, by Daniel Suarez. en.wikipedia.org/wiki/Daemon_... No, not because it’s good. In fact it is not great writing and cheesy as hell. Because the Torment Nexus Suarez describes in 2006 is being re-invented

Aya reposted

What is your University internal deadline like? Single deadline or double deadline? -CIHR LOI date Feb 4, 2026 -Dalhousie FoM internal deadline on Uni server Feb 12, 2026 -Dalhousie ORS internal deadline on Uni server & ResearchNet Feb 17, 2026 CIHR Deadline Mar 4, 2016 @dalhousieu.bsky.social

Aya reposted

13% success rates at #CIHR, is a really 87% fail rate for>4000 scientists. At a time when 🇨🇦 is investing millions to recruit more scientist into a broken funding environment. 🇨🇦, U need a strong science foundation 2 attract success. #CIHR, fix your science foundation, it is eroding in real time

Aya reposted

Replying to @andymasley.bsky.social

The article itself has a lot of great coverage of people harmed by nitrate pollution from farms. This headline is like running a story on how a cigarette company caused cancer, but the headline only mentioning an unrelated bowling alley in the town, and implying the bowling alley caused the cancer