Sneha Patel
Computer Science Student & AI/ML Developer building practical AI systems and products.
About Me
I'm Sneha — a Computer Science student at Navrachana University and an AI/ML developer focused on building practical AI systems and products. Whether it's architecting real-time educational AI for Arcanum AI or integrating edge deep learning on hardware for SustainX 2026, I focus on solving real-world challenges with rigorous engineering.
Computer Science student at Navrachana University with an active focus on Machine Learning, Deep Learning, Computer Vision, Agentic AI, and IoT edge systems. Experienced in architecting real-time AI solutions like Arcanum AI (TetraTHON 2026, 6th Rank / 160 Teams), developing physical smart waste classifiers (GreenSort / SustainX 2026 paper), and analyzing high-volume behavioral NLP datasets. Driven to build trustworthy, production-ready AI systems that solve grounded real-world problems.
Experience
Co-Founder & AI / Backend Technical Lead
Co-created Arcanum AI, an educational intelligence system designed to help students—especially international and non-native English speakers—understand complex university lectures. Spearheaded the AI/ML pipeline and asynchronous FastAPI backend architecture. Secured 6th Rank out of 160 teams at TetraTHON 2026 (Navrachana University & ISEN Méditerranée). Currently scaling from hackathon MVP toward an agentic syllabus RAG architecture.
- Built low-latency contextual concept simplification API
- Created classroom telemetry schema capturing comprehension bottlenecks
- Architected prompt routing pathways to reduce latency under live evaluation
Data Specialist Intern
Conducted data preprocessing, feature engineering, and model evaluation on a benchmark crisis text dataset containing 101,110 records. Engineered an NLP text classification workflow to assess distress sentiment patterns, and modeled contextual relationship pathways using a Neo4j graph database to move beyond isolated keyword matching.
- Preprocessed and tokenized 101,110 textual records with sentiment scoring
- Trained Scikit-Learn classification ensembles to identify distress indicators
- Constructed Neo4j graph schemas linking users, emotional markers, and distress tokens
Machine Learning Intern
Built an end-to-end car recommendation engine utilizing multi-parameter user preference scoring, feature engineering, and regression/classification models. Packaged the pipeline with a functional user interface and documented system performance metrics.
- Engineered feature normalization and categorical encoding pipelines for automotive telemetry
- Developed recommendation ranking algorithm based on user budget and feature weights
- Delivered technical documentation and validation analytics to the senior engineering team
Undergraduate AI & IoT Researcher
Co-authored research on 'GreenSort: An Integrated IoT and Deep Learning System for Real-Time Municipal Solid Waste Segregation, Hazard Detection, and Route Optimization' for SustainX 2026. Designed a custom ResNet classifier running on Raspberry Pi CPU at 1.7s latency, communicating via UART with an Arduino Uno sensor array.
- Fine-tuned 4-class ResNet waste classifier achieving 90 mAP at 1.7s edge inference
- Integrated 6 hardware sensors including ultrasonic fill and MQ gas sensors
- Validated prototype across all 8 end-to-end hardware-software test scenarios
Education
B.Tech in Computer Science & Engineering
Navrachana University, VadodaraActive Undergraduate StudentRelevant Coursework: Data Structures & Algorithms, Object-Oriented Programming, Database Management Systems, Machine Learning Foundations, Deep Learning, Computer Vision, Agentic AI, Software Engineering, AI in Healthcare.
Diploma in Computer Engineering
Govt. Polytechnic for Girls, Ahmedabad9.42 CGPA · Final Semester SPI 10.0Relevant Coursework: Computer Programming (C, C++, Java), Database Systems, Computer Networks, Operating Systems, Web Development. Graduated with 9.42 CGPA and SPI 10.0
Research & Papers
GreenSort: An Integrated IoT and Deep Learning System for Real-Time Municipal Solid Waste Segregation, Hazard Detection, and Route Optimization
Authors: Hetvi Sheth, Parisha Parmar, Sneha Patel
Venue: SustainX 2026 (Navrachana University)
Guide: Prof. Gaurav Kumar Gaharwar, Assistant Professor, CSE, Navrachana University
Municipal solid waste (MSW) generation is projected to reach 3.4 billion tonnes annually by 2050. Existing smart bin deployments focus almost exclusively on fill-level monitoring, leaving waste classification, weight-based density anomaly detection, and hazardous gas monitoring largely unaddressed. GreenSort is a prototype Smart Waste Management System (SWMS) designed for municipal deployment. An Arduino Uno multi-sensor array feeds a Raspberry Pi running a custom-trained ResNet classifier fine-tuned via transfer learning, achieving 90 mAP at 1.7s edge inference latency on CPU. Telemetry and classification tags stream via MQTT to a Supabase cloud database, driving dynamic route optimization and citizen overflow reporting. Total prototype cost: approximately ₹5,000.
DermaJourney: Multi-Modal Longitudinal Decision Support Architecture for Chronic Dermatological Care
Authors: Sneha Patel
Venue: Undergraduate Academic Research Proposal
Investigates a multi-modal decision support framework for chronic dermatological conditions. Unlike conventional single-snapshot classifiers, DermaJourney explores sequential lesion feature delta tracking coupled with clinical NLP extraction from patient-reported symptom chronologies. Focuses on clinician-in-the-loop decision support with transparent uncertainty bounds, framed strictly as academic methodology research without unverified diagnostic claims.
Achievements & Certifications
Credentials & Achievements
Certifications (5)
Neo4j Certified Professional
Issued by Neo4j GraphAcademy · 2024AWS Academy Graduate — AWS Academy Machine Learning Foundations
Issued by Amazon Web Services (AWS) · 2024Google Cloud Academy — Becoming a Google Cloud Data Engineer
Issued by Google Cloud (GCP) · 2024Honor & Achievements (5)
6th Rank out of 160 Teams — TetraTHON 2026
2026Team 'Twin Tech Girls' secured 6th Rank at TetraTHON 2026 — an Indo-French AI Innovation Hackathon hosted at Navrachana University in partnership with ISEN Méditerranée, France. Co-created Arcanum AI, leading the AI/ML pipeline and backend technical architecture.
Research Paper (Co-Author) — SustainX 2026
2026Co-authored 'GreenSort: An Integrated IoT and Deep Learning System for Real-Time Municipal Solid Waste Segregation' presenting a validated edge-AI smart bin prototype with ResNet achieving 90 mAP at 1.7s latency.
Academic Distinction — 9.42 CGPA / SPI 10.0 Topper
2024Graduated Diploma in Computer Engineering with academic distinction, securing a 9.42 CGPA and a perfect SPI 10.0 in the final semester at Govt. Polytechnic for Girls, Ahmedabad.
WSRO Nationals Robot Race Competitor
2023Represented the institution in national-level robotics championship, designing and programming an autonomous high-speed tracking robot using C++ and Arduino.
CEO Hackathon Commendation (Mesh Works)
2024Designed and built an AI Supplier Ranking Agent that analyzes defect rates, late deliveries, and cost metrics. Received a personal commendation from the CEO of Mesh Works.
I believe AI should solve grounded, real-world problems and amplify what people can achieve. My approach is simple: understand the core problem deeply, build robust architectures, validate claims with empirical data, and never stop learning.