Vipul
Thota
I build intelligent data systems and computer vision solutions that transform raw information into actionable insight. Currently at UWorld engineering ETL pipelines and AI-powered tools — with a passion for research at the frontier of deep learning.
Who I Am
A passionate engineer at the crossroads of data, AI, and computer vision — turning complex problems into elegant, high-impact solutions.
B.Tech Computer Science & Engineering
CGPA: 8.12 / 10 · 2021–2025
My journey into AI started with a fascination for how machines can perceive the world — leading me to build production-grade computer vision systems for railway safety and to publish research at Stanford. Today I engineer data pipelines at UWorld, powering learning experiences for millions of students.
What drives me
- 🔬Research-driven engineering
- 🚀Shipping products that matter
- 📊Data telling meaningful stories
- 🤝Open-source collaboration
Core Technologies
Work Experience
Building production AI systems and data pipelines at fast-moving organizations.
- Design and maintain robust ETL pipelines processing millions of student assessment records daily.
- Architect scalable data warehousing solutions using SQL, improving query performance by 40%.
- Build automated data quality monitoring frameworks to detect schema drift and anomalies.
- Collaborate with product teams to deliver analytics dashboards that inform learning-experience decisions.
- Implement CI/CD workflows for data pipeline deployment using Docker and Git-based version control.
- Developed an OpenAI-powered chatbot for student support, reducing manual query resolution time by 60%.
- Built an automated PPT generation tool using LangChain and OpenAI APIs for educator content creation.
- Engineered an AI-driven flashcard generation system from curriculum documents, processing 500+ pages.
- Integrated REST APIs with the LMS platform for seamless deployment of AI features at scale.
- Optimized prompt engineering strategies achieving 92% content relevance score in user testing.
Featured Projects
A selection of impactful engineering projects spanning computer vision, data engineering, and AI/ML systems.
Real-time detection and segmentation of railway track defects using state-of-the-art object detection models, enabling proactive maintenance before critical failures occur.
- ✦Real-time defect detection at 30 FPS
- ✦90%+ mAP across crack, joint & corrosion classes
- ✦Automated defect classification pipeline
- ✦Full image processing & reporting API
A full-stack platform for geospatial data visualization and predictive analytics, integrating multiple map providers and real-time data ingestion for decision support.
- ✦99.9% platform uptime in production
- ✦Multi-format GeoJSON & shapefile ingestion
- ✦Interactive heatmaps & choropleth layers
- ✦Predictive geospatial analytics module
An intelligent recommendation engine that suggests laptops based on user requirements and budget constraints, powered by ML models and exposed via clean REST APIs.
- ✦85% recommendation accuracy score
- ✦Dynamic multi-criteria filtering logic
- ✦Scalable microservice architecture
- ✦RESTful API with Flask & Django backends
Publications
Peer-reviewed research accepted at the International Workshop on Structural Health Monitoring (IWSHM 2025) at Stanford University.
AI-Driven Railway Maintenance for Fault Identification Through Object Detection and Segmentation
Vipul Thota et al.
This paper presents a comprehensive AI framework for automated railway fault detection using ensemble object detection (YOLO, RTDETR) and instance segmentation (Mask R-CNN), achieving 90%+ mAP on real-world rail imagery. The system enables proactive maintenance scheduling, reducing derailment risks.
Smart Railway Safety: Integrating Deep Learning with Vision Transformers for Obstacle Detection and Track Health Monitoring
Vipul Thota et al.
This work introduces a hybrid architecture combining CNN-based feature extractors with Vision Transformers (ViT) for dual-task railway monitoring — simultaneous obstacle detection and continuous track health assessment. Deployed on edge hardware, achieving real-time inference at sub-200ms latency.
Technical Skills
A well-rounded stack spanning backend engineering, computer vision, machine learning, and data infrastructure.
Backend Engineering
Computer Vision
ML / Deep Learning
Generative AI
Data Engineering
Tools & DevOps
Achievements
Recognition, competitive milestones, and academic accomplishments that mark the journey.
Awarded the Mahindra University Merit Scholarship for academic excellence in Computer Science.
Ranked in top 15% of the B.Tech Computer Science batch at Mahindra University.
Solved 150+ algorithmic problems across LeetCode and Codeforces, with strong DSA fundamentals.
Two research papers accepted at IWSHM 2025, Stanford University — both on AI for railway safety.
Services
Available for consulting, freelance projects, and research collaborations in these areas.
AI Solutions Development
- ▸ LLM-powered chatbots & agents
- ▸ RAG pipelines with LangChain
- ▸ OpenAI & Hugging Face integration
Data Engineering & ETL
- ▸ Scalable ETL pipeline design
- ▸ Data warehouse architecture
- ▸ Real-time streaming data systems
Computer Vision Systems
- ▸ Object detection & segmentation
- ▸ Custom YOLO model training
- ▸ Video analytics pipelines
Backend API Development
- ▸ REST API design with Flask/FastAPI
- ▸ Database modelling & optimization
- ▸ Containerized microservices
Let's Work Together
Whether it's a consulting engagement, research collaboration, or an exciting full-time opportunity — I'd love to hear from you.
I'm open to Data Engineering roles, AI consulting, and research partnerships. If you have a problem at the intersection of data and intelligence — let's solve it together.
- Remote Data / AI Engineering roles
- AI consulting projects
- Research collaborations in CV / MLOps
- Part-time freelance engagements