carlcarrie Profile Banner
Carl Carrie (@๐Ÿ ) Profile
Carl Carrie (@๐Ÿ )

@carlcarrie

Followers
16K
Following
353
Statuses
25K

Meanderings on markets, microstructure, misinformation, machine learning, metaverses, and multiverses... maybe more. #FinTech #AI #Python - Views are my own.

New York
Joined June 2009
Don't wanna be here? Send us removal request.
@carlcarrie
Carl Carrie (@๐Ÿ )
12 days
DeepSeek's influence on AI and Markets is profound, with its open-source approach challenging traditional moats. There are no AI MOATS! Read more here:
1
0
3
@carlcarrie
Carl Carrie (@๐Ÿ )
11 days
RT @swapnakpanda: "Foundations of Machine Learning" A 505-pages book from MIT for beginners is ๐Ÿ’ฏ FREE.
Tweet media one
0
662
0
@carlcarrie
Carl Carrie (@๐Ÿ )
12 days
Deepseek Impact across Hyperscalers, AI-Scalers, AI-Ecosystem and AI-Chip Players Discussedโ€ฆ
Tweet media one
0
0
2
@carlcarrie
Carl Carrie (@๐Ÿ )
15 days
Speech Certainty - Stanford Law Review on Algorithmic Speech vs 1st Amendment #LLM
Tweet media one
0
0
1
@carlcarrie
Carl Carrie (@๐Ÿ )
19 days
Foundations of Large Language Models New #ebook covers pre-training, generative models, prompting techniques, and alignment methods...
Tweet media one
0
1
13
@carlcarrie
Carl Carrie (@๐Ÿ )
25 days
Quantum Cats on a Saturday Morning... A new discovery promises to greatly reduce errors in quantum computing and is an important precursor to Applied Quantum Computing - paper
Tweet media one
0
0
4
@carlcarrie
Carl Carrie (@๐Ÿ )
1 month
If you ever want to read a visual, non mathematical but thorough history and description of Entropy Well, this is it...
0
1
5
@carlcarrie
Carl Carrie (@๐Ÿ )
1 month
@a16z @a16z also discuss other expected transformations in how AI will change user interfaces and compliance with regulations in FinTech/DaaS. Both reduce the value of existing Moats for DaaS / FinTech providers and potentially open new revenue lines.
Tweet media one
Tweet media two
0
1
0
@carlcarrie
Carl Carrie (@๐Ÿ )
1 month
NVIDIA just announced DIGITS, a GPU-powered Super Computer for AI that can run models with 200B parameters and daisy chain to run 400B, 600B, or higher model sizes (truly large language and multimodal models) - all this in the ultra-compact form factor of a Mac mini.
@TechCrunch
TechCrunch
1 month
At #CES2025, Nvidia CEO Jensen Huang unveiled Project Digits, a โ€œpersonal AI supercomputerโ€ that provides access to the companyโ€™s Grace Blackwell hardware platform in a compact form factor.
0
0
0
@carlcarrie
Carl Carrie (@๐Ÿ )
1 month
Hmmm Lots to unpack...
0
0
0
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
RT @docmilanfar: The origins of the formula my dad knew is a mystery, but I know it has been used in the bazaar's of Iran (and elsewhere) fโ€ฆ
0
19
0
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
RT @giansegato: this is paper is kinda wild. turns out that if you simply ask an LLM to straight out predict a timeseries like this: ``` <โ€ฆ
0
200
0
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
Large language models as inspiration for time series foundation models Workshop : List of papers:
0
1
17
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
Good discussion on entropy and the decades of different perspectives on defining it
1
4
10
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
@a16z and @ycombinator lists of sought investment archetypes in AI, FinTech, Digital Assets and tech and vertical infra
0
0
1
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
New paper integrates deep learning with rough volatility models FinX
Tweet media one
0
2
12
@carlcarrie
Carl Carrie (@๐Ÿ )
2 months
Multimodal Time-Series Research Page: โ€œOur results show that plotting as an approach amplifies the capabilities of multimodal models currently available to consumers to make sense of complicated time-series data.โ€ Paper:
Tweet media one
0
1
23