software | Rishit Dagli

In a past life, I used to do a lot of software engineering. I have listed a few open-source projects I started (apart from research codebases). I have presented a few talks at software conferences which are also listed here.

selected software

torch-diffsim

Parallelizable, differentiable, deformable simulation entirely in PyTorch

Code Project Webpage PyPI

Simplicits Nerfstudio

Qualitative, speed improvements and a more complete implementation for Simplicits which allows simulating a mesh, Gaussian Splat, or a NeRF.

Code

MIRNet-TFJS

TensorFlow JS models for MIRNet for low-light💡 image enhancement that can run entirely on your browser.

Code Video

Fast-Transformer

An optimized implementation of Additive Attention.

Code PyPI

3D Transforms

A library to easily work with 3D data and make 3D transformations.

Code Project Webpage PyPI

Gradient-Centralization

Instantly improve your training performance by implementing Gradient Centralization in optimizers.

Code PyPI

Perceiver

An optimized implementation of Perceiver.

Code PyPI

Greenathon

Originally a hackathon submission, shows how to train models specifically for deploying them to run entirely on browsers.

Code

ISAB

A framework to use Permutation-Invariant Neural Networks.

Code PyPI

ML With Android 11

Popular samples for optimized inference for machine learning models on Android using TensorFlow Lite using capabilities introduced in Android 11.

Code Video

Face-Recognition Flutter

Popular samples for optimized inference for machine learning models on Android using Flutter and Firebase ML Kit.

Code

TF Watcher

A tool to monitor your ML jobs remotely.

Code Project Webpage PyPI

Nystromformer

An optimized implementation of using Nyström Method for approximation self-attention.

Code PyPI

Transformer in Transformer

An optimized implementation of performing attention inside local patches for image classification.

Code PyPI

Conformer

One of the first implementations of the popular Conformer.

Code PyPI

GLOM

One of the first implementations of the popular Hinton's GLOM with optimization to make it runnable.

Code PyPI

GLU

Gated Linear Units and many of their variants.

Code PyPI

Compositional Attention

An optimized implementation of Compositional Attention and their variants around Disentangling Search and Retrieval.

Code PyPI

Invariant Point Attention

Invariant Point Attention from AlphaFold 2 for all problems.

Code PyPI

Wasm FAAS

A proof of concept to run Machine Learning models as serverless functions with Wasm.

Code

conference talks

About	Conference
But How Do You Autograd Through Your Simulator? Making a Differentiable Simulator in PyTorch	PyTorch Conference 2025
Orchestrating Machine Learning on Edge Devices	PyTorch Conference 2023
Building Highly Scalable Edge Computing with Project Akri & WebAssembly	Embedded Open Source Summit 2023
GitOps for Machine Learning Pipelines	CdCon + GitOpsCon 2023
Making the Most out of your Hardware Accelerators in a Kubernetes Cluster	KubeCon + CloudNative Con Europe 2023
Supercharging your Kubernetes Deployments with Wasm	KubeDay Japan 2022
Open Source, Kubernetes, And CloudNative From the Eyes Of a High-Schooler	KubeCon + CloudNative Con North America, 24 October 2022
Serverless Magic for ML Orchestration using Kubeflow	Kubeflow Summit, 18 October 2022
Educational Reference Video on WebAssembly	Webpage