Dylan
dylantartarini.bsky.social
Dylan
@dylantartarini.bsky.social
Data Scientist 👨🏻‍💻🐍

Based in Bologna 🇮🇹🇪🇺

MSc in Economics 📊

Passionate about books, mountains, politics and policies, coding, artificial intelligence and quality conversations. 📖⛰️👾
Pinned
It's been a while since my last story on @medium.com.

Had a lot of fun building my own take on the concept of "LLM Wiki" and its @obsidian.md plugin, now part of my day-by-day workflow.

pub.towardsai.net/building-my-...
Building my own LLM-Wiki Research Team
Compounding knowledge using AI Agents
pub.towardsai.net
I’ve just had my say on our future banknotes. Have your say too! surveys.ecb.europa.eu/10b/neweuro/
Survey | surveys.ecb.europa.eu
I’ve just had my say on our future banknotes. Have your say too!
surveys.ecb.europa.eu
August 3, 2026 at 5:21 PM
Reposted by Dylan
It's been a while since my last story on @medium.com.

Had a lot of fun building my own take on the concept of "LLM Wiki" and its @obsidian.md plugin, now part of my day-by-day workflow.

pub.towardsai.net/building-my-...
Building my own LLM-Wiki Research Team
Compounding knowledge using AI Agents
pub.towardsai.net
June 22, 2026 at 11:03 AM
Reposted by Dylan
The Pokémon Trading Card Game AI Battle Challenge is now live - in partnership with @Pokemon_cojp 📢

Build AI Training Agents through two connected competitions focused on strategic gameplay in the Pokémon Trading Card Game environment.
June 17, 2026 at 12:59 PM
Reposted by Dylan
Bluesky launched in February 2023 as an invite-only platform.
As of June 2026, we have 45 million users.
For reference, that's almost the population of Argentina.

Thank you for being here with us 💙
Here’s to the next 45 million accounts.
June 25, 2026 at 7:03 PM
It's been a while since my last story on @medium.com.

Had a lot of fun building my own take on the concept of "LLM Wiki" and its @obsidian.md plugin, now part of my day-by-day workflow.

pub.towardsai.net/building-my-...
Building my own LLM-Wiki Research Team
Compounding knowledge using AI Agents
pub.towardsai.net
June 22, 2026 at 11:03 AM
Reposted by Dylan
The #Gemmaverse just got more open! With Gemma 4 now under the Apache 2.0 license, we're providing the autonomy and clarity developers need to build breakthrough innovations. What does this unlock for you? #OpenSource #AI goo.gle/gemma-4-apac...
goo.gle
Google releases Gemma 4, expanding the Gemmaverse with Apache 2.0-licensed AI models. Learn how this empowers developers and researchers.
goo.gle
April 2, 2026 at 4:00 PM
Reposted by Dylan
This excellent article showcasing how to build sophisticated LLM applications by @dylantartarini.bsky.social is a prime example of the in-depth, application-focused #LLM insights that we are currently seeking.
Build and Query Knowledge Graphs with LLMs | Towards Data Science
Going from document ingestion to smart queries — all with open tools and guided setup
towardsdatascience.com
May 21, 2025 at 9:27 PM
Reposted by Dylan
👏 Introducing Dylan Tartarini 👏 Dylan joins TDS as one of our newest authors with a strong background in data science and economics.
Build and Query Knowledge Graphs with LLMs | Towards Data Science
Going from document ingestion to smart queries — all with open tools and guided setup
towardsdatascience.com
May 18, 2025 at 1:47 PM
Reposted by Dylan
⛪️ Network analysis applied to the conclave and predicting the next pope. 3 factors: status, information, and alliances.

Love it. Article here:
www.unibocconi.it/en/news/netw...
May 10, 2025 at 7:01 PM
Reposted by Dylan
Level up your RAG! Dylan Tartarini's new article shows you step-by-step how to build and query Knowledge Graphs with LLMs using open tools (Neo4j, LangChain, Ollama + more) for smarter, inter-document understanding.
Build and Query Knowledge Graphs with LLMs | Towards Data Science
Going from document ingestion to smart queries — all with open tools and guided setup
towardsdatascience.com
May 2, 2025 at 11:06 PM
Glad to have been published for the first time in Towards Data Science 😍 this is kind of a mini-dream come true!
Level up your RAG! Dylan Tartarini's new article shows you step-by-step how to build and query Knowledge Graphs with LLMs using open tools (Neo4j, LangChain, Ollama + more) for smarter, inter-document understanding.
Build and Query Knowledge Graphs with LLMs | Towards Data Science
Going from document ingestion to smart queries — all with open tools and guided setup
towardsdatascience.com
May 3, 2025 at 5:36 AM
Now, this is kinda interesting..

Makes you think that all these #AI people are focusing on social networks (for data gathering? What other purposes?)

What can be said other than "Mundus sine Caesaribus,"?

www.theverge.com/openai/64813...
OpenAI is building a social network
ChatGPT versus X?
www.theverge.com
April 15, 2025 at 4:36 PM
Reposted by Dylan
Happy to share here my very first Medium Story, freshly published. In my intentions, this is the first of a series on #KnowledgeGraphs and #GenAI.

medium.com/@tartarinidy...

#ML #AI
On Knowledge Graphs and their relevance
Going past Vector Databases to organize domain knowledge in organizations.
medium.com
March 24, 2025 at 9:37 PM
Happy to share here my very first Medium Story, freshly published. In my intentions, this is the first of a series on #KnowledgeGraphs and #GenAI.

medium.com/@tartarinidy...

#ML #AI
On Knowledge Graphs and their relevance
Going past Vector Databases to organize domain knowledge in organizations.
medium.com
March 24, 2025 at 9:37 PM
Reposted by Dylan
Questo fatto che le mie tasse pagano la ricerca pubblica ma per leggerne i risultati devo ri-pagare non riesco proprio a digerirlo

#openscience
March 11, 2025 at 10:04 AM
Reposted by Dylan
It's amazing how quickly 'network engineer' can turn into 'network villain' with a single command.
March 11, 2025 at 8:34 AM
Reposted by Dylan
📢 App Version 1.99 is rolling out now (1/5)

With every update, we’re aiming to make the Internet less toxic, more fun, and more in your control.

• 3 minute videos!
• A separate request inbox for DMs from unknown users
• Mute accounts more quickly
March 10, 2025 at 6:28 PM
Reposted by Dylan
Anyone working with Knowledge Graphs here?

Would be interested in getting in touch with people working on real-world applications based on #GenerativeAI and #KnowledgeGraphs : I am developing something on my own and I have a few technical matters I'd like to discuss.

#DataScience #RAG #AI #GenAI
March 9, 2025 at 5:05 PM
Reposted by Dylan
Il sesso è potere?
È difficile trovare parole nuove, ma proprio per questo è necessario lavorare su report come questi
ondata.substack.com/p/sesso-e-po...
📃 SESSO È POTERE?
Sì. Anche nel 2025 il potere è in mano agli uomini
ondata.substack.com
March 6, 2025 at 7:56 AM
Anyone working with Knowledge Graphs here?

Would be interested in getting in touch with people working on real-world applications based on #GenerativeAI and #KnowledgeGraphs : I am developing something on my own and I have a few technical matters I'd like to discuss.

#DataScience #RAG #AI #GenAI
March 9, 2025 at 5:05 PM
Reposted by Dylan
In 2024, the world was around 1.5°C warmer than it was in pre-industrial times. But temperatures haven’t increased linearly; there have been spikes and dips along the way.
March 7, 2025 at 11:58 AM
Reposted by Dylan
This chart is from our Spanish Instagram account: www.instagram.com/ourworldinda...

Read more about our translation efforts: ourworldindata.org/instagram-in...
Login • Instagram
Welcome back to Instagram. Sign in to check out what your friends, family & interests have been capturing & sharing around the world.
www.instagram.com
March 7, 2025 at 12:11 PM
Reposted by Dylan
Italy's data protection authority is asking Chinese artificial intelligence startup DeepSeek for formal responses on how it handles Italians' data when they use its AI chatbot app.
Italy’s privacy regulator goes after DeepSeek
Italy’s data protection authority is asking Chinese artificial intelligence startup DeepSeek for formal responses on how it handles Italians’ data when they use its AI chatbot app.…
www.politico.eu
January 29, 2025 at 12:04 PM
Some attention points here:
* to be considered open source, models should have: (1) open weights, (2) open data, (3) open architecture, (4) open code.
* OpenAI does not provide any (you can guess point 3).
* DeepSeek only provides (1) and (3).
www.ft.com/content/a0df...
OpenAI says it has evidence China’s DeepSeek used its model to train competitor
White House AI tsar David Sacks raises possibility of alleged intellectual property theft
www.ft.com
January 29, 2025 at 10:15 AM
Reposted by Dylan
“Did trillion-dollar capitalists just roll over and get in line when faced with obvious cartoonishly bald-faced fascism?”

“Probably.”
The moves drew praise from Trump, who called them "impressive" and said of Meta, "they've come a long way." Asked whether he thought they came in response to his threats — Trump once suggested prison for Zuckerberg — he responded, "Probably." wapo.st/4h223hP
Meta ends fact-checking, drawing praise from Trump
Meta CEO Mark Zuckerberg ended fact-checking on Facebook and Instagram in favor of community notes, calling the recent election a “cultural tipping point” on free speech.
wapo.st
January 8, 2025 at 3:59 AM