🙌Introducing GitHub Kudos

One thing about being a developer, is how frustrating the job search process can be. One of the things I encountered was the request for recommendations from other colleagues.

I am lately quite active in GitHub, I created some small libraries, and I also contributed to others, so, in case any recruiter checks my GitHub profile (most recruiters don't do), they can see my repositories, but there’s no straightforward way to showcase who you’ve worked with or who recommends working with you.

App GitHub Kudos

GitHub lacks a simple, personalized way to share personalized shoutouts, so I decided to build a solution: GitHub Kudos.

GitHub Kudos is an app that generates an image featuring a GitHub user's avatar with their recommendation. My intention is that these images can be displayed in GitHub profiles and/or personal websites to share how other developers recommend you (or who you recommend).

If you want to see how it looks, you can check my profile https://github.com/manuelarte#-people-i-recommend

Or if you want to see how the image looks by itself https://github-kudos.com/manuelarte/kudos/octocat

I don’t think this will completely eliminate recruiters asking for recommendations, but I hope it makes it easier to share and showcase them. 😊

How Does It Work

The app is quite simple, let's say Alice (github login alice) wants to recommend Bob (github username bob), then this are the steps:

• Alice creates a repository called github-kudos (alice/github-kudos).
• Alice creates a file bob.md in which she writes her recommendation to Bob.
• Either Alice or Bob can access the kudos image by https://github-kudos.com/alice/kudos/bob

(I created a template repository in GitHub to facilitate this setup)

Mancala and IA

🤖My First AI

I was always curious how an AI in a game works, to be honest it is quite a complex topic, and I thought it could be interesting to learn how to do an AI for a small board game.

Instead of choosing the typical Tic Tac Toe, I decided to use Mancala. Mancala is a board game originally from Jordan whose rules are not very complicated and I though it could be a good start to learn how to do my first AI.

The goal of this post is not to explain how the game Mancala works, but rather focus on how I implemented the AI algorithm to play the game. If you aren't familiar with the game, I recommend you to read its rules and play one or two games to get a feeling.

Q-Learning

There are several algorithms to create an AI, but the algorithm I will use to build the AI is Q-Learning. As a summary, Q-learning is a reinforcement learning algorithm that finds an optimal action. It helps an agent learn to maximize the total reward over time through repeated interactions with the environment, even when the model of that environment is not known.

In the Mancala's case, it will store the number of beads on every bowl and its possible actions and consequences.

How the algorithm works

Imagine the following situation (state) S1 (the player mancalas are on the sides, in the example the player 2' mancala contains 5 seeds, and the player 1' mancala contains 8 seeds. The 12 pits are placed in two rows with 6 pits each row):

 _____Player2_____________________________________________________
/  _____    ____    ____    ____    ____    ____    ____          \
/ |     |  [____]  [____]  [__1_]  [____]  [____]  [____]   ____  \
/ |  5  |                                                  |    | \
/ |_____|   ____    ____    ____    ____    ____    ____   |  8 | \
/          [____]  [____]  [____]  [____]  [____]  [__1_]  |____| \
/                                                                 \
/____________________________________________________*Player1_____\

The Player1 has only one option (action) A5. If it plays that action, then the game is finished and Player1 wins.

Then, the algorithm will save that, all previous states and actions that leads to this final state and action,makes it wins.

(S1, A5) = 1

Before this state, we had a previous one, e.g.: 

 _____Player2_____________________________________________________
/  _____    ____    ____    ____    ____    ____    ____          \
/ |     |  [____]  [____]  [____]  [__1_]  [____]  [____]   ____  \
/ |  5  |                                                  |    | \
/ |_____|   ____    ____    ____    ____    ____    ____   |  8 | \
/          [____]  [____]  [____]  [____]  [_1__]  [____]  |____| \
/                                                                 \
/____________________________________________________*Player1_____\

 In this state S2, the Player1 has one action, A4, then, if it plays that action, the game is still on, so he would save something like this:

(S2, A4) = 0

So basically it will give a score to the different states and actions.

So by playing many games (training), the knowledge (states and actions) of the algorithm increases and it's able to make intelligent decisions.

Playing a game

If you want to try it out, don't hesitate to check out my repo: https://github.com/manuelarte/mancala-go

🔗References:

• 💻Github Project
• 🥚Mancala
• 📖Q Learning Algorithm

Tarkov Profile Parsed (ToastedSherpa)

I want to learn 🐍Python🐍, and following the approach of previous posts, I decided to do a fun project for it.

To find some motivation, I though it could be a good idea to do something related to the game Escape From Tarkov, for the ones don't know, it's a shooter-looter quite challenging.

💡Idea

The problem with this game is that, there are many cheaters. Most of the Discord servers related to this game, I am in, have like a channel dedicated for people to post screenshots of profiles that are suspicious.

The profile gives some statistics about the player, like the number of raids played, or the number of players killed. Here it's an example on how my profile in 0.14 wipe looks like:

I thought it could be fun to make a Discord bot that parses that image to extract player's information, in order to check if there is like a way to "identify" patterns in those profiles.

👷Implementation

So, basically the profile picture has always the same format. On the left side you have the picture of the player and the name, and on the center and right side you have the achievements and more statistics.

We are going to focus on parsing the left side of the screenshot.

For that we are going to use OpenCV, to manipulate the image to being able to extract data from it, and Tesseract (and open-source OCR), to convert the manipulated image to text.

Parsing The Name

The name is always in the same position, at the bottom of the player's image. So we can crop that part to make it easier for the OCR.

Tesseract has problems to identify characters in a black background, so in order to facilitate that, we are going to make the image black and white, and invert the colours.

This is the code used:

def transform_name_image(cropped_image: cv2.typing.MatLike) -> cv2.typing.MatLike:

  bw_image = cv2.cvtColor(cropped_image, cv2.COLOR_RGB2GRAY)

  return cv2.bitwise_not(bw_image)

def parse_name(image) -> str:

  cropped_image = crop_name(image)

  transformed_image = transform_name_image(cropped_image)

  output = pytesseract.image_to_string(transformed_image)

  return parse_name_output(output)

With these changes, Tesseract is able to guess the name properly.

Parsing The Level

Same principle to parse the level, but in this case, we are going to tell Tesseract that it should find only numbers in the parsed image.

Parsing Extra Info

TODO

Technologies/Frameworks Used:

• 🐍 Python
• 🤖 DiscordPy
• 🖼️ OpenCV
• 👁️ Tesseract

🔗References

• 🖥 Github project (currently is private): https://github.com/manuelarte/tarkov-cheater-shamer-python
• If you are interested in learning more about this, feel free to contact me. I have a python playground where you can try it out.

Tarkov Team Kill Tracker

I want to learn 🦀Rust🦀, and for that I decided to do a fun project.

I confess I really like the game Escape From Tarkov, for the ones don't know, it's a shooter-looter quite challenging.

Here you can find a video of me, killing one of the most difficult bosses in the game, Killa:

When you play with friends, there is no way to identify who is your teammate (apart from asking questions like: is that you?, are you in <insert place>?). 

For that reason, it's quite common to have team kills. 👀

💡Idea

I though it could be a good idea to create a Discord bot to keep track of the team-kills that happens. To make it a bit more fun it could have achievements. I came up with some of them, as an example:

• 🎉First team kill registered🎉
• 👺Killed by five different friends👺
• ...

Also, to promote people using this bot, I though it could be nice if the "killer" get roasted by that team-kill. So then, once the team kill is registered, ChatGPT would reply to the command by roasting the player that made the team-kill.

👷Implementation

The flow is the following:

1. User registers who team-killed him using a discord command: > /tk @Manuelarte
2. App registers the team kill and
1. Checks for achievements
2. Requests to ChatGPT a Roast to the user.
3. Replies the message to the author plus the achievements achieved with that team-kill.

Technologies/Frameworks Used:

Here is a list of some of the libraries used for this project:

• Poise
• Diesel
• r2d2