Mayank Agarwal

Deep learning Engineer

Hello World!

I am currently building deep learning solutions at Flixstock!

Previously, I have developed quant tools to analyze data at D.E. Shaw & Co. I completed my Bachelor's in Computer Science from BITS Pilani, India with focus in Machine Learning and Data Mining. Through my undergraduate coursework, projects and internships, I have developed a great interest in Deep Learning and Computer Vision.

In my free time, I like playing badminton, watching TV shows and listening to music.

Internships

Teaching Assistant

Principles of Programming Languages
Database Systems
Object Oriented Programming

Coursework

Core - Data Structures and Algorithms, Database Systems, Object Oriented Programming, Operating Systems, Compiler Construction

Electives - Machine Learning, Data Mining, Image Processing, Information Retrieval, Neural Network and Fuzzy Logic

Independent - CS231n: Convolutional Neural Networks for Visual Recognition, Neural Networks and Deep Learning, Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization, Convolutional Neural Networks

Say Hi!

I am always open to learning, working with teams, and implementing new ideas. Let's connect on Linkedin and feel free to drop an email.

Python My weapon of choice. I've been programming in Python for over four years. I rather open an iPython console than a calculator.

PyTorch Currently building deep learning models with PyTorch. I like the data parallelism and multi gpu abstractions that comes with it.

Tensorflow Together with TFlearn and Keras, I have prior experience training deep learning models with Tensorflow.

NumPy I'm very familiar with NumPy and it's wrappers such as Pandas. I use it all the time for performing vector computations.

Ipdb Not a fan of print statements..

Touch Typing All fingers on keyboard! My favourite keys are F and J.

Linux Happy to stay close to the penguin. I love the freedom and power that comes with Linux.

Vim Not a pro user, but getting along with it!

* denotes equal contribution

3D-LMNet: Latent Embedding Matching for Accurate and Diverse 3D Point Cloud Reconstruction from a Single Image
Mayank Agarwal*, Priyanka Mandikal*, K L Navaneet*, R. Venkatesh Babu
British Machine Vision Conference (BMVC), 2018
[Code] [Video]

CAPNet: Continuous Approximation Projection for 3D Point Cloud Reconstruction Using 2D Supervision
K L Navaneet*, Priyanka Mandikal*, Mayank Agarwal, R. Venkatesh Babu
AAAI Conference on Artificial Intelligence (AAAI), 2019
[Code]