TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients.

A massively parallel, high-level programming language

Language of thought library for python 3

[ICML2024] Unified Training of Universal Time Series Forecasting Transformers

Bindings for stitch

A scalable abstraction learning library

A complete(grpc service and lib) Rust inference with multilingual embedding support. This version leverages the power of Rust for both GRPC services and as a standalone library, providing highly ef…

A one-afternoon implementation of redis-like primitives with S3 Express

Pathology Language and Image Pre-Training (PLIP) is the first vision and language foundation model for Pathology AI (Nature Medicine). PLIP is a large-scale pre-trained model that can be used to ex…

A playground for running duckdb as a stateless query engine over a data lake

Experiments and data for the paper "When and why vision-language models behave like bags-of-words, and what to do about it?" Oral @ ICLR 2023

Code associated with the paper "Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists"

[CIKM'22 Workshop(EvalRS)] Item-based Variational Auto-encoder for Fair Music Recommendation

Joining the modern data stack with the modern ML stack

Official Repository for EvalRS @ CIKM 2022: a Rounded Evaluation of Recommender Systems

Recommendations at "Reasonable Scale": joining dataOps with recSys through dbt, Merlin and Metaflow

PyTorch code for ETSformer: Exponential Smoothing Transformers for Time-series Forecasting

Code and data for the ACL 2022 paper "Probing as Quantifying Inductive Bias".

FashionCLIP is a CLIP-like model fine-tuned for the fashion domain.

Implementation of ETSformer, state of the art time-series Transformer, in Pytorch

An end-to-end implementation of intent prediction with Metaflow and other cool tools

A python package to enrich Twitter Data

Ingesting data with Pulumi, AWS lambdas and Snowflake in a scalable, fully replayable manner

Behavioral "black-box" testing for recommender systems

Materials for my 2021 NYU class on NLP and ML Systems (Master of Engineering).