Sirish Prabakar

I'm a

About

Focused and enthusiastic developer with a keen interest in software development and artificial intelligence. By comprehensive exposure to the underlying concepts and applying them vividly to various projects, my love for these domains came into being. I am a passionate individual who thrives to build and apply algorithms to solve real-world industry problems.

Location: San Diego, CA
Email: sirish999@gmail.com

Interests

Software Development

Machine Learning

Computer Vision

Natural Language Processing

Software Engineering

Visualization

Algorithms

Image Processing

Education

MS in Computer Science

2019 - 2021

Relevant Coursework

Advanced Machine learning
Software Systems and Development
Algorithms and Paradigms

B.E. in Information Science and Engineering

2015 - 2019

Relevant Coursework

Database Management Systems
Algorithms & Optimization for Big Data
Machine Learning

Online Certification

Computer Vision Nanodegree

Deep Learning Computer Vision™ CNN, OpenCV, YOLO, SSD & GANs

Experience

Software Development Engineer

Sept 2023 - Present

Apple

Played a key role in executing 100+ pre-submission tests with cross-functional teams, before software release.
Triaged over 3000 bug reports/radars facilitating software changes for high-priority bugs
Developed python/C++ scripts to streamline bug triage, email notifications for reports and log analysis
Developed XCUI-based UI automation in Swift, reducing manual testing time by 70% across multiple devices/OS.
Partnered with software and ML teams to build an automated debugging system powered by sentiment analysis

Camera Software Engineer

Feb 2022 - Aug 2023

Qualcomm.

Working in the auto exposure core team of the 3A department focusing on camera software and AI-ML assisted features
Developed a python tool for the sensor team to automatically calculate and produce different exposures of an image.
Optimized and reduced size of camera scene analyzer files used in the auto exposure solution by 52%.
Developed and implemented new touch feature in camera using object detection and segmentation network-QSegnet.

Software Developer intern

Aug 2021 - Dec 2021

GAOTek Inc.

Project 1:

Developing a Spatial AI application to assist the visually impaired using OpenCV(depth perception and object detection).
Working with QA and Dev teams and playing a key role in software development-design, implementation, and test.
Implementing the application using AWS EC2 as a server and AWS RDS Database for storage.
Automated and optimized object detection classifier freeing up to 10 labor hours/week and 7% increase in accuracy.

Project 2:

Developed a web app with BERT trained on the MIMIC-III dataset to learn the embeddings of Clinical Notes of patients.
Used the trained BERT model to predict the chances of patient readmissions to hospitals.
Built a RESTful API using the Django REST framework to generate real-time predictions from the model.
Developed a Front-end component with React to dynamically display the results of ML model on a Web App.
Designed a Back-end controller in NodeJS to fetch patient data from FHIR standard APIs of EHR vendors

Graduate Research Assistant, Machine learning

August 2021 - May 2022

California State University Sacramento

Developed a novel NLP approach to detect fake online reviews with limited labelled data and semi-supervised learning.
Classification of reviews is carried out with word embeddings and topic models in a ACGAN architecture.
Knowledge development and extensive research done in OCR, GAN, NLP, word embeddings/topics models.

Software Engineer Intern

May 2021 - Aug 2021

CalPERS

Lead a team to develop an e-Signature web application for CalPERS: California Public Employees’ Retirement System.
Conducted detailed client interviews, prioritized requirements and delivered fully functional product prototype.
Application allows employees to create profiles, extract user signatures from documents(with multiple image processing operations) and has the functionality to place it on any other company document using OpenCV library/functions.
The app saves time needed to authenticate/physically sign company documents by 40% and eliminates need for paper.
Front end: HTML/CSS/Javascript | backend: Django framework, SQLite, OpenCV in Agile environment.
Tested, designed and delivered the project 15 days ahead of schedule by efficiently designing the flow of the project.

Data Science Intern

Oct 2020 - Dec 2020

Internex Capital.

Performed data analysis and visualization with business intelligence tools such as DOMO for a FinTech company.
Created and pruned 20+ databases from unstructured customer data with functionality for automatic reporting.
Lead 2 teams to structure, debug and optimize SQL queries for mining financial and loan data from databases.