As a co-organizer of a Google Developer Group (GDG), I collaborate with a passionate team to unite the entire campus through technology and innovation. We meticulously plan and host monthly events on various technical topics, aiming not just to foster learning and collaboration, but also to reach as many students as possible. Through these efforts, we strive to make a lasting impact on our school, building a stronger, more connected, and innovative campus community.

I directed the engineering of a system, utilizing a Python-driven backend on Raspberry Pi for complex algorithmic computations and ESP32 for robust AES-encrypted communication channels. The bespoke application, intricately integrated with Apple Wallet using secure PKI standards, leverages NFC and precision RFID technologies for advanced cryptographic user authentication. This system, seamlessly integrated into expansive IoT infrastructures, facilitates MQTT protocol for efficient real-time data telemetry and employs sophisticated machine learning models for predictive security analytics, ensuring a highly secure, scalable, and intelligent access control solution for critical environments.

I am currently spearheading the development of a pioneering Real-Time Voice Transcription system at my company, aimed at revolutionizing healthcare documentation. This innovative product, leveraging Python on Raspberry Pi with cutting-edge ASR technologies such as Google’s SpeechRecognition, Mozilla’s DeepSpeech, and OpenAI's WhisperAI, is designed to enhance the efficiency of medical professionals. By accurately transcribing medical consultations in real-time and integrating seamlessly with Electronic Health Records (EHR) through robust networking protocols and Docker for scalable deployment, this system reduces manual documentation errors and saves valuable time. This solution not only aims to streamline physicians' workflows but also positions our company to expand its product line into new, technologically advanced healthcare solutions.

In my role as a mentor at CodeDay, I provided round-the-clock support and expertise during a 24-hour technology event that fosters innovation and hands-on experience among its participants. My responsibilities spanned multiple disciplines, including engineering, mechanical design, electrical systems, and software development. I guided participants through both the conceptualization and execution stages of their projects, offering critical troubleshooting support and problem-solving strategies to ensure optimal outcomes. This experience not only allowed me to contribute to an inspiring educational environment but also to assist in nurturing the next generation of tech innovators.

As a Comprehensive Volunteer (Level III), I supervise and mentor students during school and at school-sponsored activities, including overnight trips. My responsibilities include ensuring student safety, facilitating educational, robotics and recreational activities, and providing support and guidance. This role often involves working independently, without constant oversight from school staff.

•Selected by Purdue for membership based on academic standing and leadership potential•Participate in Success Networking Teams to bring leadership to Campus

This research pioneers a fusion of computer science and biology to address urgent environmentalchallenges by developing a sustainable fertilizer through composting. It aims to reduce the ecological impact of agriculture and tackle waste management issues. Using a sophisticated machine learning model, trained to discern waste types, including biodegradable materials, the project underscores a commitment to environmental sustainability and technological prowess. By segregating biodegradable waste for composting, it promotes soil health and productivity while mitigating the global landfill crisis. Aligned with Sustainable Development Goal 15 - Life on Land, this initiative addresses waste management and soil health, highlighting the potential for meaningful innovation in agriculture and waste management. Faculty: Assistant Professor Alanna Olah

The Safe Scan research project, originating from Purdue University, is a pioneering effort in computerscience and security technology. It aims to revolutionize threat detection and prevention through advanced computer vision technology and artificial intelligence (AI). By analyzing real-time visual data, Safe Scan distinguishes normal activities from behaviors indicative of security risks, such as aggressive actions or the presence of firearms. It seeks to transform traditional security measures from reactive to proactive, enhancing safety in communities, schools, businesses, and public spaces. This proactive approach minimizes the risk of violence or unrest, fostering peaceful coexistence by preventing incidents before they occur. Beyond security, Safe Scan contributes to computer science and AI by pushing the boundaries of machine learning and computer vision. It demonstrates the potential of AI-driven solutions to address societal challenges and emphasizes the importance of continuous innovation in security. In a society where violence is increasingly common, Safe Scan offers an efficient, non-intrusive method of threat detection, ensuring the safety of public spaces. In conclusion, the Safe Scan research project represents a significant advancement in creating safer environments, highlighting technology's potential to address modern security challenges through innovation. Faculty: Assistant Professor Thomas Bolinger

This research project delves into the realm of computer vision and video analysis, specifically focusingon the pivotal role of object tracking in entrance surveillance scenarios. With the objective of identifying the most reliable and efficient algorithm capable of monitoring individuals navigating entrances, a comprehensive evaluation of ten prominent object-tracking algorithms is undertaken. These algorithms, ranging from traditional methods to cutting-edge deep learning based approaches such as DeepSort, Lucas Kanade, and Boosting, are rigorously scrutinized against a custom dataset designed to emulate real-world surveillance conditions. Through a blend of quantitative metrics and qualitative analysis, including tracking accuracy, speed, and robustness to occlusions, the research aims to provide practitioners and researchers with valuable insights for selecting optimal tracking solutions. Anticipated outcomes include not only a comparative analysis of algorithm performance but also contributions to the broader understanding of surveillance system efficacy and the potential for innovation in the field. Ultimately, this endeavor seeks to empower advancements in computer vision and video analysis, with implications extending to public safety and the evolution of surveillance technologies. Faculty: Dr. Mohammadreza Hajiarbabi

Developed the Mask Detector and the Occupation Tracker for the temple. The Mask Detector uses Deep Learning models to detect if the incoming individuals are wearing their masks, if they are not wearing their masks, the Mask Detector plays a message asking them to put it on. The Occupancy Tracker keeps account of the number of people entering the temple and exiting through two different doors. Once the max threshold of the temple is reached, the Occupancy Tracker lets people entering the temple know that the temple is full and they should wait outside. Appreciation Letter from the NESSP Temple: https://tinyurl.com/NESSPAppreciationLetter, Volunteer Hours Letter: https://tinyurl.com/MaskOccupationVolunteerHours

I was selected as an Inspirit AI Leadership Fellow and took part in the AI Ambassador program taught by MIT and Stanford graduates. AI Scholars is a pre-collegiate AI education program. Please learn more at www.inspiritai.com

I worked as an organizer at Hack3, a 24-hour global event where students "hack" together solutions to today's most pressing problems, network with other students, attend workshops and receive feedback from judges.

I worked on various tasks improving the service which was offered by the company.

LPBI’s IT and Data Science problem solved: To extract the views for each of the 6,000+ articles for different time periods. Must be done in a time-efficient manner that retrieves the necessary data for LPBI. Create a multiple-step process that involves web scraping the data in multiple processes to enhance the speed of the extraction, and use data analysis libraries in Python to format the data in the correct format. IT Infrastructure Projects: LPBI 2.0’s Official Slide Deck LPBI India Map SOP Article for how to update the map of LPBI India. Finish Onboarding of LPBI SOP Manual for how to get an email on DropBox. Updated the CVs for VL, PK, IG Converted a selective number of Biographical articles of LPBI’s Founder into Podcasts Contact WordPress to figure out solutions for issues. Create documentation for Data Science Project Create a presentation for LPBI July Internal Meeting.Publication on PharmaceuticalIntelligence.com Developing Deep Learning Models (DL) for the Instant Prediction of Patients with Epilepsy Developing Deep Learning Models (DL) for Classifying Emotions through Brainwaves February 1, 2021 – We Celebrate 1,924,400 e-Readers, 6,000 Scientific Journal Articles, 7,525 Scientific Comments, A Journal Ontology of 728 Medical & Life Sciences Research Categories, 10,440 Tags, Top Article 17,300 Views, Top Author 487,500 Views on PharmaceuticalIntelligence.com Standard Operating Procedures for creating a DropBox Account Standard Operating Procedures for Updating the Map of LPBI India