GianPaolo Santopaolo

4.8K posts

GianPaolo Santopaolo banner
GianPaolo Santopaolo

GianPaolo Santopaolo

@gsantopaolo

AI Engineer - Generative Models - Agentic - Fine Tuning - RAG - MLOps - AI Clusters

Switzerland Katılım Mart 2010
364 Takip Edilen779 Takipçiler
GianPaolo Santopaolo retweetledi
Theo - t3.gg
Theo - t3.gg@theo·
Theo - t3.gg tweet media
ZXX
173
2.3K
35.4K
1.2M
GianPaolo Santopaolo
GianPaolo Santopaolo@gsantopaolo·
@MatthewBerman Did some Python/ML development with it. For simple task way slower than 4o Complex tasks, well you might be right but I see some limitations in the chain of thoughts that really look’s like an agent…
English
0
0
0
6
Matthew Berman
Matthew Berman@MatthewBerman·
OpenAI just dropped GPT Strawberry and the results are wild. It is the best model I've ever tested, it only failed ONE question (can you guess which one?) Check out these incredible examples of GPT-o1 dominating my LLM rubric. 👇
Matthew Berman tweet media
English
16
22
239
45.4K
GianPaolo Santopaolo
GianPaolo Santopaolo@gsantopaolo·
🧑‍💻 Developers, how likely are you to use an AI tool that writes all your repetitive code in your unique style? This tool would learn from your source code, internal docs, and guidelines to generate code that fits perfectly into your project. 🚀 #DevTools #AI #CodeGeneration
English
0
0
1
67
GianPaolo Santopaolo
GianPaolo Santopaolo@gsantopaolo·
CogniX semantic service running on my M2 analyzing a 1.77 MB pdf: - conversion to MD - open and chunking (fixed size atm) - creating embeddings (via SentenceTransformer) - storing the 59 entities in Milvus total time: 5.69 seconds!!! Now, let's try on the real GPU cluster.
GianPaolo Santopaolo tweet media
English
0
0
1
59
GianPaolo Santopaolo
GianPaolo Santopaolo@gsantopaolo·
CogniX, our RAG system, runs in a Kubernetes cluster. In the screenshot, the semantic service analyzing my OneDrive and storing embeddings in Milvus ready for retrieval.
GianPaolo Santopaolo tweet media
English
0
0
1
58
Lior Alexander
Lior Alexander@LiorOnAI·
GPT-Engineer just hit 12,000 stars on Github. It's an AI agent that can write an entire codebase with a prompt and learn how you want your code to look. ▸ Asks clarifying questions ▸ Generates technical spec ▸ Writes all necessary code ▸ Easy to add your own reasoning steps, modify, and experiment ▸ Lets you finish a coding project in minutes.
English
67
320
1.9K
543.6K
Flightradar24
Flightradar24@flightradar24·
Okay, avgeeks, can you tell us to which aircraft this landing gear belongs?
Flightradar24 tweet media
English
680
83
2.4K
1M
Brian Ortiz
Brian Ortiz@brianvortiz·
📢 130+ KPIs: A Handbook for SaaS Companies 📢 Measure the right SaaS KPIs with specialized examples and explanations The guide contains info in the following categories: Sales Marketing Retention Finance & more comment 'KPIs' and I'll DM you the link. (MUST be following)
Brian Ortiz tweet media
English
160
11
54
9K