I'm currently collaborating with the Vector Institue (under Prof. Animesh Garg) and NVIDIA's AI team to build a soft body API for soft object manipulation and dynamic cutting. We are developing a novel surgical robot simulator using Intuitive's da Vinci system, and training RL methods to do sub-tasks on Isaac Sim. Alongside this, we've been collaborating with researchers at Johns Hopkins University on integrating their simulators into Isaac Sim. We will be publishing more on this publicly shortly. 

I've been a huge fan of O'Reilly for as long as I can remember. The book "Introduction to Machine Learning with Python" by Andreas C. Müller and Sarah Guido was one of the very first machine learning books that I ever read. 

I'm really pumped to share that I'll be publishing my own book “Machine Learning for Robotics”, alongside co-authors Marco Mascorro (co-founder of Fellow AI) and Keerthana Gopalakrishnan (research engineer at Google AI). We will be touching upon cutting-edge methods like Gato, learning-based (RL, IL), vision, and language models for robotics.

You can stay posted on my Twitter for more frequent updates on this. An early release of our first two chapters will be out in the next month.

I had the opportunity to co-author a research paper with Hanson Robotics sharing our work with Open Arms, a low-cost open-source platform designed to extend the capabilities and accessibility of humanoid robotic grasping and manipulation. Alongside an open SDK and development environment, we present a novel Generative Grasping Residual CNN (GGR-CNN) model that can generate robust antipodal grasps with a 92.4% accuracy on new household objects. We presented an early version of this work at ICRA 2022!

The latest version of our work is now released under the General Public License and the designs are available for any developer to use freely. You can access our CAD designs here. 

🎙Presenting research at ICRA 2022!
 

🏔Retreats and Travels!
 

As much as I could, I tried to carve out time these past few months to go down curiosity rabbit holes, give myself space to learn and go on adventures.

With my friends Davide Radaelli and Mukundh Murthy, I organized a learning retreat for 10 of my friends to come together, learn about a topic of our choice and share our findings. I even booked an impromptu trip to Banff with my closest friends where we spent time hiking (...almost falling off mountains) and exploring nature. 

I started a blog series (over 15K subscribers) with Oliver Cameron where we dive into different technical topics and share a blog with our learnings! Think low-level enough to interest engineers, but high-level enough for everyone to grasp why it matters.

In our first deep dive, we talk about work that Meta Reality Labs (MRL) is doing using soft robotics, and a high-speed microfluidic processor to generate artificial touch and feel. There are many interesting applications of this in surgery, neuromuscular rehabilitation for stroke survivors, and prosthetics. 

This fall I'll be continuing my studies at UC Berkeley studying CS, with an entrance scholarship. My goal is to build my technical competency in ML, and hardware and further explore my interests in startups! Really excited to live in the Bay Area for the next few years and to meet more amazing people, researchers and founders there :)

Thanks so much for following along on my journey and being an important mentor or friend in my life. For the most part, I will be located in San Fransico for the next few months. If you're around, let's definitely catch up in person! I will also be in the following places and would love to meet up if you're around:

📍Mendocino Woodlands - Interact Fellowship Retreat from June 16-19th. 

📍Geneva, Switzerland - Villars Council Symposium from June 21st to 24th and will (most likely) be there Sept 21st - 23rd.