@_NicT_ Profile picture

Nicholas Teague

@_NicT_

Freelancer and Investor

Joined February 2009
Similar User
Animesh Garg photo

@animesh_garg

Amira Abbas photo

@AmiraMorphism

Sobalaan 🧉 photo

@Sobalaan

BowtiedCuda | Dating & Relationships photo

@BowtiedCuda

Nancy Elena Black photo

@nancyelenablack

Morgan McGuire photo

@morgymcg

Khimya photo

@khimya

Paul Vicol photo

@PaulVicol

Kunal Sharma photo

@kunalsharmaq

Amirata Ghorbani, PhD photo

@GhorbaniAmirata

Rui Wang photo

@ruiwang2uiuc

Alexandru Panici photo

@alxpr1c3

Richard Nagel photo

@RichardDNagel

Valentina Zantedeschi photo

@vzantedesc

Josh Jordan photo

@jordancurve

Someone hire this schmuck a lawyer please.

The automunge patent wasn’t an attempt to own tabular preprocessing for supervised learning and whatnot, it was awarded based on demonstrating novelty of the abstractions used as basis for the API. (The claim 1 from which other claims are dependent.) The “family tree primitives”.



Venture capital is so more fun than retail investing because rather than being driven by balance sheet algebraic minutiae the number of intangibles requiring human judgment and interpretation comes mile a minute. A stock investor may have one unique insight a year, in VC per week


Human managed systems that attempt to ensure fair and just outcomes by way of following the strict letter of the law are only fair and just if there is mechanism to escalate and fix when circumstances are identified in which the letter of the law is demonstrably unfair and unjust


I don’t know if it is widely known that when openai released version 1.0 in 2023 their chat completions option for API calls changed the default from automatic retention of context between multiple samples to requiring users to manually aggregate context between API calls.


I expect that in the future economy, as the AI / workforce rebalancing starts to accelerate, different forms of activist investing will represent a majority of those meaningful forms of knowledge work available to professionals who find themselves without a recent college degree.


An economy secures so much benefit in having a diverse field of competing firms, even in traditionally winner take all markets. Every platform will have their own innovation path and invention in one segment may not have been seen without deviations in other layers of tech stack.


LLM performance metrics are unintuitive, typical paper shows set of acronyms and decimal values incoherent to non experts. Idea: the field should create a metric that quantifies the scaling of performance improvements in regimes from zero to few shot learning. Easy to benchmark.


I don’t wish to shoehorn a specific quantum computing product to profit. I wish to use my experience with research and industry to contribute to find reasonable and appropriate ways to expedite commercialization for benefit of our country. Skin in the game helps focus attention.


The long text summarization application, which language models are already ok at, could be a good way to leverage LLM in a educational setting. Students summarize a book or something, model compares and critiques, student repeat over and iterate till they intuitively understand.


Please note that I have previously shared a url that redirected to my medium essays which is currently experiencing some registration renewal snafu, going forward the URL’s turingsquared(.com) and a few others will be allowed to expire. The authoritative links remain on LinkedIn.


Just because there will most certainly be aspects of the emerging economy and media ecosystem that will be difficult to understand for a while doesn’t mean that we should abandon those principles and values that got us here. We need to remain anchored to what defines our society.


Loading...

Something went wrong.


Something went wrong.