speech-to-intent-cognichip Ternary Edge AI chip design for speech to intent for 31 tasks for the FSC dataset. Meant for deployment on the Zynq 7020, Zybo Z7
SystemVerilog 4/19/2026
mistral-7b-v0.3-peft-tuning-sentiment140 An end-to-end pipeline for binary sentiment classification on the Sentiment140 dataset. Includes comparative analysis between LoRA and Adapters, automated hyperparameter tuning via Optuna, and instruction-tuning for Mistral-7B-v0.3. Reaches >93% F1-score with minimal VRAM footprint.
Jupyter Notebook 4/19/2026
gemma-2b-fine-tuning-dolly-15k Fine-tuning Google's Gemma 2B on the Databricks Dolly 15k dataset using Hugging Face Transformers. This project implements full parameter fine-tuning with rigorous evaluation using LLM-as-a-judge (GPT-5.2 and DeepSeek-v3.2) across In-Distribution (Dolly) and Out-of-Distribution (Alpaca) datasets.
Jupyter Notebook 4/19/2026
wixqa-rag Building and evaluating Retrieval-Augmented Generation system for the WixQA dataset with Cerebras and gpt-oss-120b. We evaluate different chunking and retrieval strategies k={1,3,5,10}, chunks={300,500}, overlap={30,60}.
Jupyter Notebook 4/18/2026
bencejdanko.github.io
HTML 4/18/2026
bencejdanko
4/17/2026
sign-language-translation video-to-text sign language translation with a modified SAM-3D-Body library fork, evaluated on How2Sign
Jupyter Notebook 4/17/2026
cube-standard Standardize benchmark wrapping so the community can wrap various otherwise-incompatible benchmarks uniformly and use them everywhere.
4/7/2026
tldraw-astro A robust integration for rendering tldraw diagrams in Astro projects using standard Markdown directives.
HTML 4/3/2026
voxtral-live-agent-agx-orin Scripts to create a live, low-latency and low memory voice-agent system using the AGX Orin
Python 3/27/2026
m55m1-ElevatorCounting-YOLOv8n People-counting from top-down elevator views. Using the Nuvoton M55M1 and its Ethos-U55 NPU, the system runs optimized ML models directly on the device.
Python 3/24/2026
prepare-overhead-person-detection In total, these are 13,469 images and annotations for overhead head-count detection in confined spaces. Prepared for the elevator-people counting project for Nuvoton.
Python 3/22/2026
mamba Mamba SSM architecture (forked for AGX Orin compilation). With 130M params, we achieve 225 tokens per second, which is constant regardless of context length, due to space-state architecture.
Python 3/20/2026
imagenet-classification-mobilint-mla-100 ImageNet (subset) Classification on Mobilint's MLA-100 Edge NPU
Jupyter Notebook 3/18/2026
llm-tabular-titanic Using 5 advanced prompting techniques as an LLM-based prediction system to predict whether a passenger survived the Titanic disaster.
Jupyter Notebook 3/16/2026
protein-function-prediction scripts for extracting different protein-sequence embeddings for the kaggle CAFA-6 competition (2025)
Jupyter Notebook 3/14/2026
how2sign-rgb-front-clips This repository contains some preparation scripts for uploading structured front-view clips from How2Sign to HuggingFace.
Python 3/14/2026
manifests Kubernetes manifest files to easily self host at scale with ArgoCD. Includes configs for n8n, langflow, langfuse, loki + otel, mcp forge, getOutline, argo workflows, superset and more
3/7/2026
bert-tweeteval BERT is an encoder-transformer based architecture that is more suitable for some tasks, such as classification. We benchmark both base and fine-tuned performance trained on the TweetEval dataset.
Jupyter Notebook 3/6/2026
n-gram-modeling Comparing a Count Trigram model and Neural Trigram model after training on speeches from Thomas Jefferson.
Jupyter Notebook 3/6/2026
cadence.md Notion clone attempt, a WYSIWYG editor, slash web deployment platform created from notes
TypeScript 3/5/2026
thorn Automated agentic research intelligence system — discovers papers on arXiv, builds a self-growing knowledge graph, tracks trends and novelty, generates polished markdown digests, and pushes summaries to Discord.
Python 3/5/2026
kanjivg-stroke-classification
Jupyter Notebook 3/5/2026
solona-graduation-ml This repository contains Google Colab notebooks and modeling code for the Solana Skill Sprint: Memecoin Graduation competition (ended April 30th, 2025).
Jupyter Notebook 3/1/2026
openweather-airflow-dbt A data pipeline project that utilizes Apache Airflow and dbt to extract, transform, and analyze weather data from OpenWeather API.
Python 2/18/2026
inclusive_world_portal A full-stack role-based management portal for the Inclusive World team.
Python 2/12/2026
compiler-evaluation-server
Python 2/5/2026
StartupIdeaTrackerWeb
2/3/2026
kong-plugin-jwt-claims-headers This plugin enhances the standard Kong JWT authentication by automatically adding claims found within a validated JWT to the headers of the request sent to the upstream service.
Lua 1/24/2026
mla100-npu
Python 11/21/2025
cloudflare-url-unfurler
TypeScript 7/1/2025
uber-eats-clone-v1
JavaScript 5/22/2025
uber-simulation-system
JavaScript 5/12/2025
network-rails-flink-kafka This a Flink-Kafka stack and pipeline setup for analyzing the Network Rails feed.
Python 4/25/2025
quine-redpanda-cassandra-cypherql
Shell 4/24/2025
flink-window-dgim-traffic
Java 4/23/2025
exp-dec-window-apache-pinot
Python 4/21/2025
MAAI-Hackathon GMG - MAAI Hackathon Event. Small Streamlit, Monte Carlo, vector database experiments.
Python 4/3/2025
card-transactions-kafka-flink-mongo A mock pipeline to simulate a stream of credit card transactions. Kafka, Flink and MongoDB are used for real time analysis and storage.
Python 4/2/2025
bloom-flajolet-implementations Python implementations and simulation results for analyzing P-Vector Bloom Filters and the Flajolet-Martin Algorithm.
Python 4/2/2025
spotify-with-lyrics-dataset-generator A set of helper scripts to combine song attributes provided by Spotify to lyrics sourcable on LyricGenius.
Python 4/2/2025
text-streams-kafka-spark-hive A simulation of n number of incoming concurrent textual streams. Inputs are stored and parsed in an ML pipeline in real time. Meant to simulate LLM streams (Kafka, Spark, Hive)
Python 4/2/2025
flashcards-crud-client A basic CRUD client application for managing flashcard decks and corresponding questions and answers. (React, ShadCN)
TypeScript 4/2/2025
GeoAttendance Users within geographical boundaries will be able to check-in (Browser Geolocation API). Upon checking-in, the users will receive an email which confirms their attendance status. (React, Pocketbase)
JavaScript 4/2/2025
pocketbase-api-template A PocketBase API template—Golang, JavaScript hooks, and lightweight deployment.
JavaScript 4/2/2025
uber-eats-clone-v0 Comprehensive Uber Eats clone powered by React, Redux Toolkit/RTK, Node, MySQL, S3, Sequelize– food ordering with restaurant management.
TypeScript 4/2/2025
solitaire-scrabble A casual vocabulary game. A dynamic programming algorithm analyzes the optimal word sequences. (Python, Flask, Jinja2)
Python 4/2/2025
markdown-transpiler Markdown transpilers: state machines and plugins (Typescript, Javascript, Deno)
JavaScript 4/2/2025
carillo Car Dealership CMS experiment (SvelteKit, React Native w/ Expo, PocketBase)
JavaScript 4/2/2025
K-R
Zig 5/9/2024
MancalaProject A mancala board game developed with OO and MVC principles | Fall 2022 Object Oriented Course Project
Java 2/19/2024
CourseProject A storage and order system for music media | Winter 2022 Data Structures Course Project
Java 2/20/2023