Shreya Shankar

gpt3-sandbox. The goal of this project is to enable users to create cool web demos using the newly released OpenAI GPT-3 API with just a few lines of Python.

create-ml-app. Template Makefile for ML projects in Python.

datasets-for-good. List of datasets to apply stats/machine learning/technology to the world of social good.

toy-ml-pipeline. Toy example of an applied ML pipeline for me to experiment with MLOps tools.

overleave. Chrome extension that opens and syncs Overleaf compiled pdfs in a new window.

m1-setup. Notes on how I set up my new M1 MacBook Pro

debugging-ml-talk. Code accompanying the "Debugging machine learning in production" talk

web3-reading-list. List of good readings on web3.

spade-experiments. Experiments to assess SPADE on different LLM pipelines.

ml-dataval-tutorial. Tutorial: Data Validation for Machine Learning Techniques

char-rnn-tensorflow. Multi-layer Recurrent Neural Networks (LSTM, RNN) for character-level language models in Python using Tensorflow

planner. A "smart" planner that determines when to study, work on assignments, etc.

oreilly-monitoring. Jupyter Notebook

100-exercises-to-learn-rust. A self-paced course to learn Rust, one exercise at a time.

research-ideas. List of proposed abstracts I'd love to work on, if I had the time.

vython. Versioning Python scripts.

datatracker. WIP experimental project to make for a better ML development UX.

shreyashankar.

questions. Questions I have that I would love to explore if I have time.

lotus. LOTUS: fast and easy LLM-powered data processing

mltrace-ifc-demo. Project demo for CS294 Privacy-Preserving Systems.

ChainForge. An open-source visual programming environment for battle-testing prompts to LLMs.

open_lth. A repository in preparation for open-sourcing lottery ticket hypothesis code.

anxiety-extension. Chrome extension for logging moods.

Personal-Website. A simple personal website template for learning web development.

prompteng. Experiment scaffold for trying out different prompts.

tensorflow. Computation using data flow graphs for scalable machine learning

motion-sigmod-demo. SIGMOD Demo 2024 Submission for Motion

wakeupnow. Detect when people fall asleep at the wheel and wake them up

streams. STREAMS: A Benchmark of Naturalistic Streaming Data for Online Continual Learning

tensorflow-wavenet. A TensorFlow implementation of DeepMind's WaveNet paper

modal-client. Python client library for Modal

holo-alfa. A minimalist, mobile first Jekyll theme.

jekyll-blog. A content-first, sliding sidebar theme for Jekyll.

verifiers. Verifiers for LLM Reinforcement Learning

cleverhans. An adversarial example library for constructing attacks, building defenses, and benchmarking both

imagenet-stubs. A teeny tiny set of ImageNet-like images for testing pipelines

prompt-eval-recommendation. Streamlit app for recommending eval functions using prompt diffs

needle-in-the-real-world. Jupyter Notebook

reasoning-optimizer. A system for agentic LLM-powered data processing and ETL

palimpzest. A System for Optimized Semantic Computation

white-paper. Simple, elegant and clean jekyll theme.

midi-rnn. Generate monophonic melodies with machine learning using a basic LSTM RNN

Wit-Facebook. Wit-Faebook Messenger Example

healthy-eating. Python

news-classifier-ui. UI for News Classifier

gcn. Implementation of Graph Convolutional Networks in TensorFlow

social-good-106-section. Workshop materials and other resources for Stanford CS 106B/X Social Good section

sentiment. Python

stylin. Java