neilc

It seems like there is an emerging consensus that (a) DL development requires access to massive compute, but (b) if you’re only using off-the-shelf PyTorch or TensorFlow, moving your model from your personal development environment to a cluster or cloud setting is too difficult — it is easy to spend most of your time managing infrastructure rather than developing models. At Determined AI, we’ve spent the last few years building an open source DL training platform that tries to make that process a lot simpler (https://github.com/determined-ai/determined), but I think it's fair to say that this is still very much an open space and an important problem. Curious to take a look at Grid AI and see how it compares to other tools in the space -- some other alternatives include Kubeflow, Polyaxon, and Spell AI.

H1B holders can actually receive social security benefits. See https://www.ssa.gov/international/agreements_overview.html

What aspect of WeWork was/is fraudulent?

I think changing the list representation made sense in part because the rest of the codebase didn't use lists in a deeply Lispy way, for the most part.

You can still see the Lisp heritage in a few places, but overall I wouldn't say Postgres is "literally lisp in C".

https://repository.law.umich.edu/cgi/viewcontent.cgi?article...

We build software to make deep learning engineers dramatically more productive. Founded by leading experts in distributed systems and machine learning, our product is a complete ML environment that enables teams of DL engineers to share a GPU cluster, do efficient hyperparameter tuning, and train and deploy deep neural networks up to 100x faster than with state-of-the-art tools. We are well-funded and have significant customer traction.

The team is deeply technical and working at the cutting edge of modern ML infrastructure, but we also care about building practical technology that solves real problems for our customers.

We’re hiring exceptional software engineers to play a fundamental role in building the company and shaping the culture. We're particularly interested in

(1) Distributed Systems Engineers who are passionate about building scalable, reliable software.

(2) Full-stack Engineers who believe that building enterprise software doesn't mean that great UX is optional.

(3) ML Engineers with a strong algorithmic and statistical background, as well as hands-on experience building deep learning applications.

Tech Stack: Python, Go, Elm, Kubernetes, Mesos, Docker, TensorFlow/Keras/PyTorch.

Compensation: Competitive salary and benefits, meaningful equity stake.

https://determined.ai/careers/

neil@determined.ai

We build software to make deep learning engineers dramatically more productive. Founded by leading experts in distributed systems and machine learning, our product is a complete ML environment that enables teams of DL engineers to share a GPU cluster, do efficient hyperparameter tuning, and train and deploy deep neural networks up to 100x faster than with state-of-the-art tools. We are well-funded and have significant customer traction.

The team is deeply technical and working at the cutting edge of modern ML infrastructure, but we also care about building practical technology that solves real problems for our customers.

We’re hiring exceptional software engineers to play a fundamental role in building the company and shaping the culture. We're particularly interested in

(1) Distributed Systems Engineers who are passionate about building scalable, reliable software.

(2) Full-stack Engineers who believe that building enterprise software doesn't mean that great UX is optional.

(3) ML Engineers with a strong algorithmic and statistical background, as well as hands-on experience building deep learning applications.

Tech Stack: Python, Go, Elm, Kubernetes, Mesos, Docker, TensorFlow/Keras/PyTorch.

Compensation: Competitive salary and benefits, meaningful equity stake.

https://determined.ai/careers/

neil@determined.ai

We build software to make deep learning engineers dramatically more productive. Founded by leading experts in distributed systems and machine learning, our product is a complete ML environment that enables teams of DL engineers to share a GPU cluster, do efficient hyperparameter tuning, and train/deploy deep neural networks up to 100x more quickly than with state-of-the-art tools. We are well-funded and have significant customer traction.

The team is deeply technical and working at the cutting edge of modern ML infrastructure, but we also care about building practical technology that solves real problems for our customers.

We’re hiring exceptional software engineers to play a fundamental role in building the company and shaping the culture. We're particularly interested in

(1) Distributed Systems Engineers who are passionate about building scalable, reliable software.

(2) Full-stack Engineers who believe that building enterprise software doesn't mean that great UX is optional.

(3) ML Engineers with a strong algorithmic and statistical background, as well as hands-on experience building deep learning applications.

Tech Stack: Python, Go, Elm, Kubernetes, Mesos, Docker, TensorFlow/Keras/PyTorch.

Compensation: Competitive salary and benefits, meaningful equity stake.

https://determined.ai/careers/

neil@determined.ai

https://www.reuters.com/article/us-eu-apple-taxavoidance/ire...

I'd be very surprised if an Amazon worker making, say, $160k consumes more in government services than the sales/property/income taxes they are paying.

This is actually not the case -- modern linkers (e.g., lld) can take advantage of multiple cores.

https://lld.llvm.org/

Disclaimer: I worked at Truviso; I know the PipelineDB guys but I'm not affiliated with the company.

It is similar, but I think an important difference is that Uber, as an organization, was deliberately flouting the law and also building tools to avoid suffering the consequences of that behavior.

I'd say it is similar to the difference between an individual using a radar detector for personal travel vs. a trucking/shipping company equipping all of their vehicles with radar detectors and making it a company policy to violate the speed limit.

https://www.postgresql.org/docs/9.6/static/sql-notify.html

(Early versions of Go had a "network channel" concept, but that was removed from the language: https://softwareengineering.stackexchange.com/questions/1540...)

> The results [of a previous study], published in October 2014, showed that some patients developed problems that adversely affected their day-to-day lives, such as difficulty driving at night or in sunshine. But it was such a small number — less than 1 percent — of the patients in the study.

> On Wednesday, the group released a follow-up report in JAMA Ophthalmology that provides more sobering information. The study suggests that the percentage of people who undergo LASIK and wind up with new visual symptoms — such as double images, glare, halos or starbursts — may be much higher.

> One interesting component of this study was that the survey showed that the percentage of people with symptoms may be much higher than what has been previously reported in studies involving direct interviews with health-care professionals.

BTW, I'm curious which jurisdictions you're referring to.

This is not true; in fact, Wells is somewhat famous for reviewing more blue-collar restaurants, in contrast with previous Times critics. For example, here is his review of Guy Fieri's restaurant in Times Square: http://www.nytimes.com/2012/11/14/dining/reviews/restaurant-...

This is not true in general: there are circumstances where Postgres can skip updating indexes when a tuple is updated, assuming the update only changes values in non-indexed columns. See:

http://pgsql.tapoueh.org/site/html/misc/hot.html

https://git.postgresql.org/gitweb/?p=postgresql.git;a=blob;f...

It's not "MBA-speak"; this sense of the word "secular" is used in economics, time-series analysis, astronomy, etc.

https://en.wikipedia.org/wiki/Secular_variation