iiinova — where your digital self keeps evolving

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 26/11/2025

Zeta Alpha launches custom AI agents built for knowledge-intensive enterprise teams. Easily create, share, schedule and manage your own AI agents that get real work done.

143

Dinos Papakostas @din0s.me · 26/09/2025

Quick thread on the EmbeddingGemma technical report. It's a pretty nice paper so check it out!

020

Dinos Papakostas @din0s.me · 18/09/2025

Really proud of this work, check it out if you're curious about the problems one typically encounters with shipping RAG solutions to production in an enterprise setting!

020

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 08/08/2025

AI agents are only as useful as the tools they can reliably access, and the latest release of our Agents SDK makes it easy to connect to MCP servers. We've prepared a quick guide where we bootstrap a minimal MCP-powered chat agent with just a few lines of code: www.zeta-alpha.com/post/build-m...

033

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 09/07/2025

Join us for the Zeta Alpha "Trends in AI" show on Friday, July 11th at 8 AM PST / 5 PM CEST live from LAB42 in Amsterdam, and online in San Francisco and around the globe. Register to receive your Zoom webinar invitation, or watch our live stream on YouTube and LinkedIn: lu.ma/trends-in-ai...

162

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 30/06/2025

In our latest blog post, we covered how we use RAGElo, an open-source toolkit we've developed internally in Zeta Alpha, to compare multiple RAG systems head-to-head and aggregate pairwise preferences into a robust, easy-to-interpret Elo ranking. Check it out: www.zeta-alpha.com/post/robust-...

zeta-alpha.com

Robust evaluations for RAG with RAGElo

Retrieval-Augmented Generation (RAG) systems have gained strong traction because of their ability to ground generated answers in knowledge sources, allowing for higher accuracy and reliability. Howeve...

011

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 24/04/2025

📢 Zeta Alpha is going to OpenSearchCon Europe 2025 📢 We are excited to be contributing back to the OpenSearch community, sharing insights from our work in Enterprise AI, Neural Search, and Retrieval-Augmented Generation (RAG). Learn more about our talks at the conference below: 🧵

132

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 01/04/2025

Join us for our monthly Trends in AI webinar on Thursday, April 3rd, at 8 AM PST / 5 PM CEST! This edition comes live from Hannover Messe, the world's leading industrial tech trade fair, and online in Amsterdam, San Francisco, and everywhere else in the world. Sign up on Luma with the link below:

lu.ma

Trends in AI: April 2025 - by Zeta Alpha · Zoom · Luma

Join us for the Zeta Alpha Trends in AI webinar on Thursday, April 3rd, at 8 AM PST / 5 PM CEST (time to be confirmed). This edition comes live from Hannover…

173

Dinos Papakostas @din0s.me · 26/03/2025

Grok starting the month with >50% is ridiculous. Did people really think Big Tech wouldn't roll out even a single incremental update? (let alone a big push like Gemini)

010

Dinos Papakostas @din0s.me · 22/03/2025

yup, it also caught my eye because they say they did this "in contrast with gecko", and although there's no ablation study in the paper on this im sure they had some internally

000

Dinos Papakostas @din0s.me · 20/03/2025

I wish they elaborated more on the model souping part though, it still surprises me to this day that weight merging just works 🧙🏻‍♂️

000

Dinos Papakostas @din0s.me · 20/03/2025

A deep dive into the Gemini Embedding technical report - turns out you don't need any fancy training objectives or architectural tweaks to get SOTA on MMTEB, just a solid base model and high quality data!

120

Dinos Papakostas @din0s.me · 14/02/2025

Wrote a little blog piece on reasoning models as a summary of our latest webinar episode, feel free to share with your non-terminally online friends who want a gentle intro to catch up and what they mean for overall AI capabilities!

020

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 04/02/2025

The whole world is buzzing with the performance of DeepSeek’s reasoning model, R1, and its implications for the dominance of Western tech companies in the AGI race. But what exactly are reasoning models? Did anyone see them coming? And how can they be used to create value?

121

Dinos Papakostas @din0s.me · 06/01/2025

sell me on ghostty? why should one switch over iterm2/kitty

110

Dinos Papakostas @din0s.me · 13/12/2024

i hope 2025 is the year of good ui/ux for ai

030

Dinos Papakostas @din0s.me · 12/12/2024

friends don’t let friends build RAG pipelines without understanding IR theory. sounds like a solid christmas gift for someone you care about!

031

Dinos Papakostas @din0s.me · 12/12/2024

Drowning in the 4000 main track papers at #NeurIPS? Here’s a guide to 10 research areas and some of the spotlight papers that grabbed my... attention!

110

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 11/12/2024

It's #NeurIPS week! With 4K+ main papers, it's OK to feel overwhelmed. Our annual NeurIPS guide is here to help! We've curated a list of 10 topics and papers to check out, whether you're attending the conference or want to stay updated on the biggest developments in AI.

142

Dinos Papakostas @din0s.me · 07/12/2024

wait what happened, why do i have apple intelligence on my mac... 👀

030

Dinos Papakostas @din0s.me · 06/12/2024

the real reason we don't have agi yet is that uv's documentation isn't in the llms' training data

010

Dinos Papakostas @din0s.me · 05/12/2024

"think harder" is going to become in 2025 what "shorter" has been in 2024

000

Dinos Papakostas @din0s.me · 05/12/2024

on a similar note, this is another banger paper you should read

000

Dinos Papakostas @din0s.me · 05/12/2024

how about you o1-pro vide some benchmark numbers and real use cases

020

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 05/12/2024

NeurIPS, the most prestigious conference in AI, kicks off next week! We will guide you through the top themes and highlights, tomorrow in our webinar, and next week with a detailed blog post and video. We picked 5 papers that we previously covered in our webinar which will be presented as orals: 🧵

151

Reposted by Dinos Papakostas

Zeta Alpha @zeta-alpha.bsky.social · 05/12/2024

Join us this Friday for the Trends in AI webinar, as we reflect on the breakthroughs that defined 2024 and see how our predictions held up. Plus a NeurIPS highlight reel, plus the trending AI papers of the month, like OpenScholar, M3DocRAG, DeMo, and more! lu.ma/oefbjkjl

042

Dinos Papakostas @din0s.me · 04/12/2024

spent the day reading the OpenScholar and Tülu 3 papers in detail. AI2 x UW has been on fire lately, this is what real “Open AI” looks like 😉. big thanks to both author teams for such thorough and well-crafted work.

140

Dinos Papakostas @din0s.me · 04/12/2024

sadly the unique content / news freshness on this platform fell off sharply after the hf dataset drama

000

Dinos Papakostas @din0s.me · 30/11/2024

happy 2 years of agi for those that celebrate

010

Dinos Papakostas @din0s.me · 28/11/2024

wouldn't mind trying it out

000

Dinos Papakostas @din0s.me · 28/11/2024

in this case (since I'm being quoted) the drama started in the ai/ml community which came to bsky to explicitly *avoid* Elon's BS with deboosting and political abuse on X, not to support him

000

Dinos Papakostas @din0s.me · 28/11/2024

yeah, it’s wild how easily people give up control of their feeds without even knowing how often that blocklist gets updated or who’s on it

010

Dinos Papakostas @din0s.me · 28/11/2024

whichever one has the best discovery feed is going to win imo, the good thing here is that the api looks extremely hackable so I hope people build custom recsys on top of it

000

Dinos Papakostas @din0s.me · 28/11/2024

i think the definitiveness of the moderation team's actions and the way they set a precedent here are the bigger issues bsky.app/profile/appu...

110

Dinos Papakostas @din0s.me · 28/11/2024

i can’t speak for him, and i totally get how it could’ve been engagement bait, but he might’ve also just been disappointed that the hf person got bullied into taking down the dataset and wanted to restore it because of that.

110

Dinos Papakostas @din0s.me · 28/11/2024

fully agree with both of your points. x is becoming more of an anti-woke propaganda machine, but it doesn't mean we need to offset that by turning this place into a decentralized cancel mob

010

Dinos Papakostas @din0s.me · 28/11/2024

i think private labs are doing everything they can to gatekeep them, good evals are the real moat

010

Dinos Papakostas @din0s.me · 28/11/2024

obviously related to the bsky firehose data export, but i don't see why doubling down on that is regarded as trolling (and why they need to take such a severe action for a service the protocol openly provides)

030

Dinos Papakostas @din0s.me · 28/11/2024

yep it'll be sad to see everything get undone and people retreating back to x just because ml folks get harassed for being pro-ai

100

Dinos Papakostas @din0s.me · 28/11/2024

never said elon got the mandate of heaven, but you can't not be critical of people who are happy to take your platform away as soon as they disagree with your views

160

Dinos Papakostas @din0s.me · 28/11/2024

i guess elon was right about anti-censorship huh? imagine getting permabanned for posting a hf dataset

7142

Dinos Papakostas @din0s.me · 28/11/2024

i guess mute/block lists should only be from people you fully trust, otherwise you're partially giving up control of your feed

020