Software Engineer Machine Learning

ABHISHEK SINGH

+1-219-***-**** Hammond, Indiana, USA

********@******.*** linkedin.com/abhishek-singh github.com/abhis201 SUMMARY

Research Assistant at the Advanced Intelligence Software Lab, PNW, specializing in public safety research. Proficient in machine learning, deep learning, and software development, with expertise in PyTorch, Computer Vision, Scikit- learn, diverse model architectures, and web development frameworks including Next.js, React.js, and Express.js. Previous experience includes 1 year as an Associate Software Engineer at Scope Retail. EDUCATION

Master of Science, Computer Science, Purdue University Northwest Aug 2023 - May 2025 Relevant Coursework: Applied Machine Learning, Artificial Intelligence, Statistics for ML, Software Engineering. GPA: 3.8/4

Bachelor of Engineering, Computer Science, Vishwakarma Institute of Information Technology 2016 - 2020 Relevant Coursework: Data Structures and Algorithms, Operating Systems, DBMS, Computer Networks. Cumulative-GPA: 8.2/10

SKILLS

Technical Skills Spring, REST, Kafka, MERN, Computer Vision, Scikit-learn, Pytorch, CNN, GAN Other Skills SQL, Git, Confluence, Jira, Agile - Sprint, Scrum Soft Skills Problem Solving, Collaboration, Critical Thinking, Adaptability Languages Java, JavaScript, Typescript, Python, C, C++ EXPERIENCE

Software Engineer for AI training data May 2024 - June 2024 Outlier.ai Remote

• Worked on improving Large Language Model’s responses by fine-tuning its training data

• Re-wrote bad responses generated by LLMs that had complex code snippets. Each response and prompt had a different scenario and a different use case.

• Worked on multiple programming languages like Java, Python and Javascript. Research Assistant Jan 2024 - May 2024

Purdue University Northwest Hammond, Indiana

• Curated a dataset of more than 500 satellite images and remote sensing data from schools impacted by mass shooting incidents. Employed advanced Deep Learning and CNN methodologies to develop a predictive model capable of assessing the risk of mass shootings at new sites by integrating multi-modal data sources.

• Implemented a CNN-based deep learning model tailored for detecting rail track cracks and vulnerabilities in real-time through videos and images, enabling proactive threat identification to mitigate potential accidents.

• Developed the proprietary Gunshot Detection System’s Software using React.js and Express.js, facilitating suc- cessful deployment across three main buildings at Purdue University Northwest, with plans for expansion to all Purdue campuses.

Associate Software Engineer Nov 2022 - Aug 2023

ScopeSys Innovation Pvt. Ltd Pune, India

• Played a pivotal role as a core team member in the development of COSMOS, the flagship Order Management System catering to retail clients across the United States.

• Collaborated on the design and implementation of Uniti, an innovative inventory tracking system intricately linked to the COSMOS platform, enhancing operational efficiency for warehouses, stores, and associated entities, seamlessly integrated with the COSMOS ecosystem for streamlined operations.

• Actively contributed to the advancement of both projects by making over 500 significant contributions to their GitHub repositories, demonstrating a commitment to continuous improvement and excellence in software devel- opment.

• Utilized diverse tools and technologies including SpringBoot, Java 17, SQL, Kafka, Cassandra, Docker, and Kubernetes to architect robust solutions and drive innovation within the COSMOS and Uniti projects. PROJECTS

RailVision - developed a cutting-edge model leveraging the YOLOv8 architecture, trained on a custom dataset comprising over 3000 images. The model accurately detects defects such as rail track cracks, broken fishplates, and faulty components, classifying them into nine distinct categories in real-time. This proactive approach aims to prevent accidents in railcars by enabling timely intervention. The project was showcased to a national railway company, receiving positive feedback and recognition for its potential in enhancing railway safety. (Try it here) SentinelSafe is a groundbreaking project aimed at enhancing school safety by developing a comprehensive risk assessment model. Utilizing a dataset compiled from approximately 50 schools affected by mass shooting incidents, the project incorporates over 300 satellite images and 500 tabular data entries. The model, built on a combined architecture of CNN and deep neural networks, integrates multi-modal inputs to assess the risk of mass shooting incidents at specific school sites. By leveraging advanced machine learning techniques, it aims to provide proactive measures for mitigating potential threats and safeguarding school environments. (Try it here) PNW Campus Navigator - powered by React.js and SpringBoot, offers live event notifications, parking spot finder, navigation, visitor guide, class locator, place search, weather alerts, and more, enhancing the Purdue Northwest campus experience, it’s deployed on Vercel and AWS. (Try it here) SharePrompt - developed with Next.js, MongoDB, NextAuth, and TailwindCSS, is a sleek AI prompt- sharing web app featuring a modern glassmorphism design. Users can manage AI prompts, access personal profiles, and utilize quick links to popular AI apps via the navbar. It offers functionalities such as ”Copy to Clipboard,” tag-based search, and password-less secure Google login. Designed for responsiveness, it ensures a seamless user experience on any device. (Try it here)

ACHIEVEMENTS

• Ranked 88th globally in Codechef Starter 62 competition.

• Achieved 3-star coder status on Codechef.

• Certified in Java by Hackerrank.

• Earned Udemy certification in Spring Boot and Hibernate.

• Received the Nurturer of the Month Award at Toastmasters Club. LEADERSHIP

• Teaching Assistant, Purdue University Northwest (PNW) Guided and instructed approximately 50 undergraduate students in Java programming, fostering their under- standing of core concepts and best practices. Conducted workshops to provide hands-on experience with version control using Git and GitHub, empowering students with essential skills for collaborative software development.

Contact this candidate