avilay

34 ms·

31.
Show HN: Torchutils – a PyTorch library for quick and systematic experimentation
2 points by avilay 6y ago|1 comments
32.
by avilay 6y ago
Does anybody else think this is in response to Apple's privacy crackdown on third party tracking data? If you buy on Buy, then you are still first party for Google and they can still get your signals.
33.
by avilay 6y ago
And given that a majority of "Ex-Google" engineers are males, where does that leave a woman (or any under-represented group) founder? She is not asking for money "just because" she is a woman. She is asking for a fair ch
34.
by avilay 6y ago
Even if her business idea was bad, or her app didn't have traction, or her LI profile was not impressive enough, she deserved, at the very least, the following: * Access to VCs, she had to use her husband's email to get access
35.
by avilay 6y ago
What do folks here think of Panda3D ( https://www.panda3d.org )? Its just a game engine without any IDE. So the typical workflow would be to build your assets and scene in Blender (or Maya, etc.) and then code up the game mechanic
36.
by avilay 6y ago
That was a pretty cool video! I loved how he used 3D Monte-Carlo and side-stepped all the complicated math. I wonder if he was using a Physics engine like Bullet or ODE to calculate the simulated trajectories. Also, a good intermediate step
37.
by avilay 6y ago
Spurred by a post on AskHN about a dev who has "fallen out of love" with programming, I wrote down my thoughts on being on the IC path as opposed to the management path. Would love to know your experiences around this issue.
38.
I still love programming – what's wrong with me?(avilay.rocks)
3 points by avilay 6y ago|1 comments
39.
by avilay 6y ago
A lot of good advice here, adding my recipe for avoiding burnout (~15yrs into my career): Delineate between not enjoying your job vs. not enjoying your profession. Sometimes getting a new job in a better work environment and especially with
40.
Ask HN: How do you decide to release a new ML model in production?
2 points by avilay 7y ago|0 comments
41.
by avilay 7y ago
The question I am trying to examine is whether it is possible to do both at once - e.g., can I learn Haskell while using it in a work project, especially when I don't know functional programming?
42.
Is “Learning by Doing” a Myth?(avilay.rocks)
2 points by avilay 7y ago|2 comments
43.
by avilay 7y ago
Nice. I have been using gitlab CI and a simple Python script to publish my notes (usually in markdown and Jupyter notebooks) to Gitlab pages. You can find the output on http://avilay.gitlab.io/gyan . And if you want to reuse
44.
Tutorial on Backprop and Compute Graphs(youtube.com)
1 points by avilay 8y ago|0 comments
45.
by avilay 9y ago
Thanks everybody for your responses. Food for thought :-)
46.
by avilay 9y ago
What is the advantage of using evolutionary algos in this case over using something like Thompson Sampling or Contextual Multi-Armed Bandit? ( http://www.kdd.org/kdd2017/papers/view/an-efficient-bandit-a... )
47.
Calling Python from React in an Electron App(gitlab.com)
1 points by avilay 10y ago|0 comments
48.
by avilay 11y ago
After moving to Azure Websites for hosting, I started developing my web app in Win10 and so far it has worked great for me. My app stack is - * ReactJS * Flask * SQLAlchemy * SQL Azure My dev tools are - * Atom for JS dev * PyCharm for serv
49.
by avilay 11y ago
Nope.
50.
by avilay 11y ago
I had built a command line tool for a similar purpose to generate REST APIs running on the MEAN stack. It creates "user" and "account" resources by default with user and admin authn/authz built in. It then deploys t
51.
by avilay 11y ago
I have to do this everytime I have to deploy, which is similar to having to create a deb package everytime Nylas has to deploy. There are no git dependencies in the process I describe above. The pip drawback that is discussed in the post is
52.
by avilay 11y ago
Here is the process I use for smallish services - 1. Create a python package using setup.py 2. Upload the resulting .tar.gz file to a central location 3. Download to prod nodes and run pip3 install <packagename>.tar.gz Rolling ba
53.
by avilay 13y ago
Thanks anatari for providing this feedback. You are spot on - our biggest technical challenge is to make a generic recommender that will work with a variety of different verticals. Our current approach is to limit ourselves to 3 specific ve
54.
by avilay 13y ago
mattmanser - thanks for the encouragement. Our website is indeed www.informion.com (flyinglizard got that right). Our service is actually just string and duct tape (a hodge-podge of Java and Python programs running on data sets contained in
55.
by avilay 13y ago
Yes, I agree that this will not have much of an effect on our chances at YC. Our objective was not to get on the front page of HN. It was to have a conversation around the pros/cons of applying for "ivy league" accelerators.
56.
by avilay 13y ago
Thanks! Fixed it. Please let us know if you spot any other typos.
57.
by avilay 13y ago
I used to be a dev who transitioned to a PM and now am the CTO in my own startup. My main reason for becoming a PM was to get more customer contact and be in a position to make customer-focused product decisions. Someone mentioned that PMs
58.
Why I left Microsoft(avilay.wordpress.com)
1 points by avilay 13y ago|0 comments
59.
What does a PM do?(avilay.wordpress.com)
1 points by avilay 13y ago|0 comments
60.
3 Things To Look For In A Startup Idea(avilay.wordpress.com)
1 points by avilay 13y ago|0 comments

More ›