Yippie! I made the firmware and its own wifi portal for setting the time and alarm, plus the button sequence needed to stop the buzzer. Lets gooo!

I finished making this bot

It now has:

/ash-ping,

/ash-roast,

/ash-dms,

plus the 67 reaction and random woof replies. The DM system was probably the most fun part to build.

Main page + custom cursor!!!

I finished the main page for my personal website!!! :D

I reused the same custom mouse CSS from my projects page, and I also used my favorite CSS that i learned from a friend. The main page has my Mii, a little about me, my projects link, GitHub, and a San Jose clock! Im done!!!! Oh also ig i made a readme too

I made the css and code for my projects page!!! :D

heyo guys! I basically added a bunch of details, like speaker holes, fillets, lid, charging port holes, etc.

I added a lot of details and heat insert holes :D

Final update/devlog !!!!

I finally finished integrating the AI pronunciation system into Anukrama.

This ended up being way more complicated than I originally thought. The model itself was trained in PyTorch using MFCCs, spectral features, and DTW comparisons against multiple good chanting recordings. The problem was that the .pt model couldn’t just be dropped into a website, so I had to export everything to ONNX and basically rebuild the preprocessing pipeline in JavaScript so the browser could generate the same kind of audio features from a recording.

I also changed how I stored the scaler data by putting the normalization values into the .npz files, then wrote the JavaScript to load the model, scaler, reference audio, process the recording, calculate the features, and pass everything through the ONNX model. There were a bunch of stupid issues along the way too, including the ONNX runtime conflicting with a variable in my JS and the scaler data not loading correctly lol.

After all of that I added the actual score display to the site, so the recording can go through the whole pipeline and come back as a pronunciation score out of 100.

This took way longer than I expected lol, but I finally got it working.

This is prob the end of this project!!!!

The Gray Thief Devlog

Started working on my new project… The Gray Thief!!!

The idea is a grayscale stealth game where you play as a thief that steals light from the world. The more light you steal, the brighter you get, which also makes you easier for guards to see.

So far I’ve:

• Made the Godot project
• Added basic WASD movement
• Added walls and collisions
• Made the player have a light value
• Made the player visually get brighter/darker
• Made walls store light
• Added light transfer between the player and walls
• Added a bounce effect when hitting walls
• Added particles when hitting walls
• Made the particles match the player’s brightness
• Drew and added the sucker/vacuum
• Made the sucker follow and rotate toward the mouse
• Made the sucker change brightness with the player
• Fixed some blurry pixel art issues
• Rounded the player slightly without using a shader

Right now I’m working on making the sucker actually interact with walls instead of just going straight through them.

Still a LOT to do, but im happy with it so far, it feels like something i can make

made a basic case :)

Devlog # 6

hiya! so this devlog i basically added stuff to the scene tree and made the styles for the DobbleScene :)

Nothing much other than that. Here are some of the photos.

i routed the pcb :)

sudo wakeup Devlog

Today I worked on the planning and schematic for sudo wakeup :)

The idea is a digital alarm clock where, when the alarm goes off, you have to find a bug in some code to turn it off.

I had to improvise to fit all 9 buttons since I only have 6 pins for them, so I’m using a 3x3 button matrix with diodes.

I also figured out the TFT connections and started laying everything out in KiCad.

The buttons will be 3 on top, 3 on the left, and 3 on the right :)

final bid battle devlog

today i finished up the css for bid battle :)

• added gradient backgrounds for the player 1 red theme and player 2 blue theme
• improved the overall look of the game
• made the buttons, cards, and item slots look better
• improved the game mode selection screen
• polished the ai loading screen
• cleaned up the layout and made everything feel more finished

and that’s it for bid battle! :)

bid battle devlog

today i added ai to bid battle :)

• added ai-generated themes and items
• added an ai loading screen + progress bar
• added ai and premade game modes
• added a bunch of ai fallbacks
• improved the bidding system
• added the 5 item limit + automatic $1 distribution
• improved error handling

next devlog i’ll work on improving the css and making everything look better :)

AI Pronunciation Model

I worked on the AI pronunciation system for Anukrama!

• I used PyTorch to train models that compare your chanting to good reference recordings and give u a score out of 100.
• I made models for Verse 1 and Verse 2 using things like MFCCs and DTW.
• The models are trained, but I need way more data because there isn’t really a big dataset of Bhagavad Gita chanting recordings.
• I still need to record a lot of the data myself.
• I haven’t added the AI to the website yet.

If u wanna help me record data for the model, pls DM me (@aashvik)!

Bid Battle Devlog #1

ok so i started working on my new game called Bid Battle

the idea is pretty simple:

• there are 2 players
• $20 each
• 10 items
• 5 item slots each

the game gives you a theme like “build your own supercar” and then you bid on different items that fit the theme.

every item starts at $1 and you type in how much you want to bid.

i wanted the AI to NOT judge the builds. it’s only going to run locally in the browser and generate the theme + items.

right now i finished the html + css and ill make the js next