If standard approach does not work => time to learn how to work hard:

[1] For every position you look for people at LinkedIn in that company. If you are already connected ask for referral.

[2] If not, look for common connections that can introduce you.

[3] If there is none => send request for adding everyone from the company to the friends.

[4] Message everyone, inviting for a coffee or virtual chat to learn about the company.

[5] If you believe that you are a good fit => say this.

----

And you spend first 4 hours of every day messaging, messaging, messaging. The response rate will be low. But you need only one job.

----

And if you do it for 6 months every day, there is no way you will not get many interviews.

----

Every interview you fail => you extensively study to address limitations of your skillset.

-----

Basically, there is no: "I have a degree, hence I deserve a job", but there is: "hard work is the goal".

-----

P.S. Somehow blog post reminds me why online dating is not working that well for men. Competition is enormous, the number of ladies is limited and things like: "I am an average man, hence I deserve attention from ladies" does not work as well

Does it work better than inputing the same information to ChatGPT?

On a parallel note. I did not find good engine for the python package documentation, built such an engine myself: https://albumentations.ai/

> They wanted it. They paid for it. They enjoyed it. The counter example is open-source software.

If we talk about popular packages: - people want it - people enjoy it - people do not pay for that

But force-feeding with strict licenses like Ultralytics does works. Yes, it is force-feeding, but noone wants to pay the price, unless there is no other choice.

If there is a strong demand for a feature, regardless of the source of the request - good enough reason to add it.

Do it in a very similar way in Cursor:

Steps: [1] Add pre-commit hook [2] Write design doc as mdc file to .cursor/rules [3] Iterate on design doc till it describes what I want [4] Ask to write the code [5] Where possible - extend the test suite. [6] On every commit check that pre-commit hook checks and tests pass [7] On every bug extend the test suite [8] Write as many as possible custom pre-commit hooks [9] Add extensive docstrings to the complex code -> adds extra context to the LLM [10] Iterate [11] From time to time ask to verify that design doc is up to date

I am all for intuitive interfaces, but I am also a big proponent of learning hot keys in every program I work with.

It would be better for design to be intuitive, but you struggle only the first time, while interfaces overloaded with information will take some of your attention every time you look at them

Interesting representation. Not rows in the database as samples and columns as features, but a whole graph.

Makes training much more flexible, and fine tuning as well. Now, when a new data in terms of samples or new tables are connected to the olds ones you just extend the existing graph, without changing its existing morphology much.

Although it is unclear if it is scientific: "Look how cool we can do" or business result: "Look how much value do we get from this representation"

Really nice.

Useful by itself, but I suspect that will get even more adoption in industry in the telemetry as details statistics about users is ultra valuable.

Thanks a lot, ternaus really appreciate your comment I’m glad you found it useful. Totally agree about the telemetry and detailed stats that was actually one of the main directions I had in mind when creating the library, which is why I added JSON format support to make it easier to send the data to servers.

"it is desirable for the software to run as a locally installed executable on your device, rather than a tab in a web browser."

An OS agnostic apps, meaning web apps is such a killer feature. You can use the same app on: Linux, Mac OS, Windows, Android, iOS

Even more, developing for web is typically faster. You made the change in the code => you see the result on the screen. For example: phone apps written in swift could be faster than ones written in react-native, but it is so annoying waiting for the compilation to finish after every small change.

----

When I worked on imagery data in an autonomous vehicles company product managers pushed us to explore the data in the cloud and it was soooo inconvenient.

As the result, PMs were ignored and everyone had a personal desktop with GPUs and fast SSDs that had the local copy of the data, so that debugging, prototyping would be fast.

As lag that one gets working with a heavy data remotely reminded moving back from SSD to slow HDD, where you needed to wait some time to see the result on the screen.

It was only half a second every time, but felt ultra annoying.