Harish Rithish

Bioinformatics and Machine Learning Researcher for Sustainability

About Me Projects Just for fun Industry Exp. Hobbies

Resume

About Me

For the past two years, I have been trying to understand the Earth's ecological cycle and the intricate connections our human economy has with it. I believe it is critical in the next few decades that we transform this deteriorating symbiotic relationship into one that is mutually beneficial. This motivates me towards building long-term solutions that address climate change. To develop the requisite skillset, I am pursuing a Masters in Computer Science at UC San Diego with a Bioinformatics specialization. Here, I am researching the regulatory genomics of green algae with Prof. Stephen Mayfield, whose biology if understood would bring us a step closer to a sustainable world.

Following my Bachelors in Computer Science from NIT Trichy (India), I spent two years as a Research Fellow with Prof. C.V. Jawahar and Prof. Manmohan Chandraker at IIIT Hyderabad (India). Here, I investigated computer vision techniques for road scene understanding (an integral component of self-driving cars) and also built a prototype to monitor traffic violations efficiently. Besides academia, I have had short stints in the industry as a Software Development Engineer / Intern at Amazon and a few startups. When I don't have my 'work-face' on, I spend the time playing football, reading Indian and Global politics or watching sports :)

Projects

Regulatory genomics of green algae

Ongoing

We are developing a motif prediction pipeline to identify Transcription Factor Binding Sites in green algae. We have identified ∼2000 potential binding sites, analyzing their genomics and RNA-seq data.
Image credits: Duygu Koca

Road Scene Understanding

Publication 2021 Demo

We proposed a novel approach that combines ground and satellite imagery to infer road scene elements such as lanes and intersections. Our work was published at ICRA, a leading robotics conference.

Automated Road Monitoring

Publication 2021 Demo

We developed a scalable road monitoring system for detecting traffic violations and road irregularities (potholes, street light absence etc.) in crowded areas through recent computer vision techniques.

Neural Question Generation

Code 2018 asds

An explorative NLP project where I developed an end-to-end trainable, attention-based sequence-to-sequence learning model that generates questions from text passages.
Image credits: Microsoft Technet

Human Fall Detection

Code 2016 Demo

Elderly people tend to slip and fall in homes and remain unattended for a long time. The proposed solution automatically detects humans falling, through cameras installed in homes, and immediately alerts family members.
Image credits: shutterstock