lysecret

39 ms·

481.
by lysecret 5y ago
Veronica decides to die. Amazing book about what it means to be crazy. How to live an authentic life. Written in a funny and easy to read way. And it has the most important quality of any book IL really like. It's short and to the poin
482.
by lysecret 5y ago
I use it to now and then reset the Dev DB. So it is marginally more convenient than running on your pc (since it takes 20 mins to reset the db). But im sure there are better usecases haha. Setting all my infrastructure in Terraform in gener
483.
by lysecret 5y ago
The main thing i learned in my first production db work is that you pay for every bit of complexity you introduce. As a noob you are somehow much more afraid to not "model the data right" so you tend to create much to many tables
484.
by lysecret 5y ago
True, its a pareto distribution as much as any creative endeavor.
485.
by lysecret 5y ago
Exactly that. Also in my expereince as a Data Science consultant was that by far the best approach was to just rebuild what they attempted in excel in python. Also now that I think of it Excel is the original low code solution haha
486.
by lysecret 5y ago
There is a space for low code solutions but you have to know the tradeoffs. I think low code belongs to initial MVPs you know you will have to re-build. Low code wont scale. Low code at some point will be an absolute pain to iterate on. Low
487.
by lysecret 5y ago
He makes more an analogy of quantum woo and decentralization woo, and i kind of agree with that. Honestly, this idea that you can pinpoint 1 similarity in a very humanly important thing and a technincal complicated, precise and solvable pro
488.
by lysecret 5y ago
our infra is running on ECS because we set it up right before docker on lambda haha. Now we dont have the time to switch. Would be much better for us though.
489.
by lysecret 5y ago
hey sounds pretty interesting can you send me your website?
490.
by lysecret 5y ago
Hey for a long time i was also very sceptical. However i can refer you to this paper to a really cool applciaiton. https://www.youtube.com/watch?v=kP-dXK9JEhY . They baseically use clever GPT-3 prompting to create a dataset,
491.
by lysecret 5y ago
Hey so quick question is there such a thing as a "modular" CRM. I am thinking of using Retool for it.
492.
by lysecret 5y ago
if you have some questions hit me up! :)
493.
by lysecret 5y ago
I mean yes, but you could say take a sufficiently large set of hidden units (say 10-100) with all their parameters and they could easily represent the kinds of functions you are talking about. So the comparrisonn is off (we probably need ab
494.
by lysecret 5y ago
I see some python / pandas hate here and I have to defend it a bit: We all agree a query like: tb_1 = pd.read_sql(big_table_1) tb_2 = pd.read_sql(big_table_2) table_i_want = pd.merge(tb_1, tb_2) table_i_want =table_i_want[table_i_want
495.
by lysecret 5y ago
Love retool I used it daily :)
496.
by lysecret 5y ago
1. "Easier to scale" Scaling is such a weired topic, its like that joke about teenage sex, everybody always talks and is concerned about it but how many people really have to scale? And generally if you have to scale you have mone
497.
by lysecret 5y ago
Yea I get that but I have experience people working with all sorts of insanely complicated heuristic to get something like a NER system running when they could have much more easily used a Hugging face model. But I totally agree that the ar
498.
by lysecret 5y ago
To me that is just an iteration on first you makes it run then you make it right. And to make it run you start by the simplest approach. And building your own model is generally not the simplest however, it can be. There are some areas wher
499.
by lysecret 5y ago
Yea i share your sentiment (dont like the spying part but hey :D). It is the most crucial skill today to be able to deal with the constant distraciton. Supplying a sheltered environment cant be the solution.
500.
by lysecret 5y ago
because it had to talk to a database which was in a private VPC, and at the same time make http requests.(in my original setup i changed that :D)
501.
by lysecret 5y ago
used to work in insurance and heavily used it.
502.
by lysecret 5y ago
Hehe used to do R IMO you are right about ggplot but I strongly disagree about pandas. I fing love it. Would love to understand you troubles with it though, after using it for 4 years daily mabye I can offer some perspective ;)
503.
by lysecret 5y ago
Excellent library for train_test_split. Jokes aside. This next to Numpy, Pandas Jupyter and Matplotlib + the DL libraries are the reason Python is the powerhouse it is for Data Science.
504.
by lysecret 5y ago
Ok im going to admit to a mistake revolving around NAT gateways and Lambdas. So, i basically wanted to connect a Lambda to a Postgres / RDS database, for that I had to put into a private VPC, but the lambdas still had to talk to the w
505.
by lysecret 5y ago
I don't know why anyone would use say mongodb when postgresjson exists.
506.
by lysecret 5y ago
I love jupyter notebooks but I think the way they have be to be used is in a "throw away" fashion. E.g. use it to explore data, develop some algorithm, then put it to PY files. It is there to develop something which is worth versi
507.
by lysecret 5y ago
a big learning for me was that people who always call themselves rational tend to me the most emotional haha
508.
by lysecret 5y ago
omg. I have a econ background and i used to read marginalrevolution religiously. Actually until economistsview shut down :( Ahhh and they mention EconTalk at the start. I think its time to dive into my past haha
509.
by lysecret 5y ago
Fraud detection as well as Facebook haha
510.
by lysecret 5y ago
Generally means graphs as either input or output or both. Discrete values and relationships aren't an issue and can be easily dealt with using embeddings.

More ›