Ok Twitter, who has an invite to Comet by Perplexity?.

This video features so many amazing people — @sergialmar @oleg_nenashev @ddobrin Dr. Christian Schustzer @JulienTopcu — that I don't deserve to be on the cover!.

OK, hear me out! Spring AI's MCP Server API is just amazing! It is so nice to use! But! I'm trying to build a native image of a local-only (studio) MCP server, and I want to build a native image from it. And it builds, without any arcane magic! But doesn't work — the list of.

RT @AlecStapp: Heat kills ~175,000 people in Europe each year and yet they still make it difficult to install AC in buildings because degro….

Woah, this is one amazing code generator! I don't like Lombok in general, but Builder here (especially staged builder) in conjunction with records is a-maze-ing!.

Write something if you wanna look at the code.I'm so grateful to myself for saving this message! Trafigura, I'm ready now! (15/15).

The solution is called "Zobrist Hash" (. It took me like 10 minutes to implement it and now without any further optimizations like memoization the whole solution takes like one minute. (14/15).

And yesterday it dawned on me that probably humanity has a solution for this exact problem and 2 minutes of googling of course gave me the solution. It is so elegant, so nice. I don't know what stopped me from finding it back then. SPOILER ALERT (13/15).

when I know other languages or can imagine a better approach to the task. At the time the hardest thing for me was to store all the boards: they all barely fit 4 gigabyes. And I needed to have them all to count the amount of unique boards matching the criteria. (12/15).

When I solved the task at that time (I wasn't any good at algorithms and was like middle-ish level developer, I think) the solution with -Xmx4g in Java took more than 2 hours to run.I saved this letter to be able to return to the task when I know more, (11/15).

They also gave a couple of examples just to give me idea how to check my solution. (10/15).

please provide with your answer the total number of unique configurations for a 6x9 board with 2 Kings, 1 Queen, 1 Bishop, 1 Rook and 1 Knight. Your solution will need to be able to run on a notebook having 4G RAM. (9/15).

the program should yield the number of distinct layouts for which all of the pieces can be placed on the board without threatening each other.When returning your solution, (8/15).

and that there are no pawns among the pieces.Write a program which takes as input:- The dimensions of the board: M, N.- The number of pieces of each type (King, Queen, Bishop, Rook and Knight) to try and place on the output, (7/15).

The problem is to find all distinct layouts of a set of normal chess pieces on a chess board with dimensions M×N where none of the pieces is in a position to take any of the others. Assume the colour of the piece does not matter, (6/15).

and C++ - languages in which those times it was common to solve algorithmic tasks.4. They didn't really care about the time of execution. At lease not at cost of the code clarity, readability, and overall maintainability.The task was the following: (5/15).

1. The solution should fit in 4 GB RAM.2. They didn't care about how long it takes to solve the task: the code quality was more important.3. They preferred not main-stream languages: Java, C, (4/15).

Usually I think that either the task is stupid, or the company is nitpicking, or it doesn't really know what it wants. However, it's not the case here. The knew exactly what they wanted: (3/15).

I store this one not because of its sentimental value. The thing is I solved the assignment but wasn't happy with my solution. I think they were not happy with it either: I had an interview round later, but they decided to stop the process. (2/15).