Avatar

Palash Chauhan

Masters Student in Computer Science

University of California, San Diego

Biography

I am a Master's student in the Computer Science department at University of California, San Diego. My interests include Distributed Systems, Machine Learning and the intersection of the two. I was a Software Developer in the Adobe Media Optimizer team at Adobe where I built a streaming architeture for search advertising attribution. I graduated from IIT Kanpur in 2017 with a major in Computer Science and Engineering. In the summer of 2016, I interned at Adobe BigData Experience Lab where I worked on integrating a Topic Model based prediction system within Adobe Illustrator.

My hobbies include reading and playing all kinds of sports - Badminton, TT, Tennis, Squash, Basketball. I was a member of the Badminton team at IIT Kanpur and won several tournaments for the institute.

Interests

  • Distributed Systems
  • Machine Learning
  • Data Engineering

Education

  • MS in Computer Science, 2021

    University of California, San Diego

  • B.Tech. in Computer Science, 2017

    Indian Institute of Technology, Kanpur

Experience

 
 
 
 
 

Member of Technical Staff

Adobe Inc

Jul 2017 – Aug 2019 Bangalore, India
Part of Adobe Media Optimizer team:

  • Designed and implemented a cross-datacenter, near real-time streaming architecture using Kafka for search advertising events attribution. The architecture improved latency by 10x.
  • Implemented a data pipeline for search keyword data using PostgreSQL, Hadoop and Presto
  • Extended AMO back-end framework for high volume and time sensitive data synchronization between AMO infrastructure and the Pinterest Ad Platform using APIs written in Python
 
 
 
 
 

Research Intern

Adobe Big Data Experience Lab

May 2016 – Jul 2016 Bangalore, India
  • Analyzed topical behaviour of users when interacting with Adobe apps like Photoshop and Illustrator.
  • Modelled user activity data using topic models like Latent Dirichlet Allocation to extract topics.
  • Used extracted topics to predict user’s intended workflow and built a recommender system based on the topic transitions to surface contextual guidance within the app.
  • Integrated the model and a prediction pipeline within Adobe Illustrator for a demo.
 
 
 
 
 

Software Development Intern

Monet Networks

May 2015 – Jul 2015 Gurgaon, India
  • Developed new engagement metrics like peak-end ratio for non verbal cues analytics and integrated them within Monet's platform.
  • Implemented a Collaborative Filtering based video recommendation system within Monet’s platform to improve user experience.
  • Enhanced Monet’s non-verbal cues analytics platform using web development in PHP, MySQL and JavaScript.

Projects

*

Benchmarking Druid and Presto

In recent years, the proliferation of internet technology has created a surge in machine generated events. These events generally have …

SurfStore

Getting people to agree on something is hard. Getting machines connected through an asychronous network to agree on something is …

Comparing Cloud Models

A comparative study of Virtual Machines, Containers and Serverless

Load Value Prediction

Predicting values loaded by machine intructions to aid Instruction Level Parallelism.

Poisson Matrix Factorization

Probabilistic Models for Bayesion Recommender Systems.

Dense Image Captioning

Teaching a computer vision system to localize and describe salient regions in images in natural language

Automatic Abstract Generation

Text Summarization for Long Documents like research papers.

Patents

Application Tool Recommendation

Computing application range significantly in their perceived complexity. Some applications are immediately intuitive. Most of the users …

Contact