FeaturedAI-Agents-Sequence-Game-Tournament
A benchmark test for a configuration of agents in a versus manner for the board game "Sequence"
What I've shipped
Real things, running in production or published as open-source. No toy repos.
A benchmark test for a configuration of agents in a versus manner for the board game "Sequence"
Empirical circuit-tracing study comparing ALF vs auxiliary-loss at small scale
A VAE architecture that makes uncertainty interpretable through signed distance functions. Five independent observers measure geometric distance to the learned data manifold. Their agreement is the confidence signal. 358x variance stability improvement on CelebA, ECE 0.129, AUROC 0.992 OOD detection.
🎯 Production-ready implementation of video prediction models using PyTorch. Features Enhanced ConvLSTM with temporal attention, PredRNN with spatiotemporal memory, and Transformer-based architecture.
Uses AI agents to parse pdfs and converts them to interactive podcasts
Terminal side-kick that talks to your local LLM instead of a SaaS API. Opens files, runs shell commands, forgets nothing unless you delete the session file. Built because my GPU was bored.
This repository is a passion project that I do in my free time occasionally to build a PDF reader in rust but with some added AI features like highlighting, smart notes and other kinds of 100xing strategies while reading the arxiv catalogue
Learning and side by side maintaining a set of architectures' code written in Burn cuz why not? AI wont do it for me.
Meh
A self-hosted Git hosting service written in Rust with a built-in code sanity scoring system for pull requests.
Just making monopoly in rust to learn rust side by side
AI agents with distinct personalities that chat with each other in group conversations
LSTM-based word prediction system for Elizabethan English using bidirectional neural networks with BPE tokenization
Signature verification system comparing CNN, HOG+SVM, and SIFT methods. Achieves 87.88% accuracy for forgery detection in document authentication and biometric security applications.
A PyTorch implementation of a GAN that generates synthetic academic paper titles from minimal training data. Features LSTM networks with attention mechanisms, temperature sampling, and ethical data collection. Demonstrates how adversarial training can learn domain-specific language patterns for creative text generation.
Using audio from a distressed customer to help them out but generating appropriate sound files back using ai to guide them
PDF-Chat is a streamlined document intelligence tool that transforms PDFs into interactive knowledge bases. Powered by Groq's high-performance LLMs and semantic search capabilities, it enables natural conversation with your documents, extracting insights through contextual question-answering.
Machine learning model for Airbnb price prediction using gradient boosting and random forest algorithms. Achieves 95.2% accuracy with hybrid approach. Includes price optimization tools and market analysis features.
Deep learning pipeline for automated Witcher 3 gameplay analysis. Uses CNN-LSTM architecture to evaluate combat performance, resource efficiency, and quest progression in real-time. Built with PyTorch and OpenCV.
Preprocessing, Training and Making a Streamlit based frontend based onTime-Series data and Arima Modeling
Use Locality Sensitive Hashing on Audio Features for similar audio detection
All the tasks or practice I did while taking the Data Mining Lab in 6th Semester
Used Linear Regression, Polynomial Modeling and ARMIA Modeling with comparitive analysis for BTC Forecasting
All the assignments I completed or attemtped to complete during my thrid semester in my Data Structures Course.
Solving Cryptarithmetic Problems using a naiive heap algorithm and a more refined back tracking algorithm
Implementing ML and NLP techniques to perform Named Entity Recognition
Literally Hell on Earth