Ashwin Santh

Exploring machine learning and automation in large-scale systems, with emphasis on fault detection, predictive modeling, and intelligent system architectures.

View Projects Contact

About Me

My Engineering Philosophy

I approach engineering like a layered system-each idea tested, visualized, and refined. I learn by building, and I build to understand. Whether it's reverse engineering malware, modeling memory in AI, or designing drone coordination logic, I treat every project as a sandbox for exploration.

I value modularity, reproducibility, and visual clarity. My tools are technical, but my motivation is intuitive: I enjoy solving problems, connecting dots, and making systems smarter. I’m not confined to one domain, I follow what’s interesting and I document what matters.

Featured Projects

Drone Decision Support System

AI/ML • Azure • Kubernetes

Designed modular ML system with multi-sensor telemetry analysis and visualization for predictive diagnostics. Demonstrated scalable fault detection and prioritization logic relevant to automation and sensor fusion in industrial systems.

Read more →

Azure Intelligent Chatbot

Azure • Docker • RAG • CI/CD

Developed generative chatbot with containerized NLP models and RAG pipelines. Built AI-native interface for structured data parsing and visualization demonstrating secure ML integration into diagnostic workflows.

Read more →

GearboxGuard

Python • FastAPI • Kubernetes • MongoDB

Built ML pipeline for gearbox fault detection using vibration signals, feature extraction, and Random Forest classification. Deployed containerized microservices with Kubernetes autoscaling demonstrating predictive maintenance in automation systems.

Read more →

AI-Powered Event Scheduler

Python • DeepSeek AI • SerpAPI • Pipedream

Developed Gen-AI agent with natural language interface for structured input parsing and alarm state modeling. Implemented threshold-based anomaly detection and webhook-based scheduling illustrating automation in regression test orchestration.

Read more →

Advanced Multi-Drone System

ROS • Gazebo • AI Research

Co-authored research on multi-drone coordination in ROS/Gazebo simulators using ML models for anomaly detection. Demonstrated 30% faster task completion and improved coordination efficiency in simulated disaster-response scenarios.

Read more →

FASHBuddy

Python • Bing API • Image Processing

Comprehensive fashion recommendation system that analyzes body shapes, fitness levels, and provides personalized fashion advice. Features facial shape detection, shoulder/waistline analysis, fitness comparison, and Bing image search integration for tailored clothing recommendations.

Read more →

Education

Master's Qualification Programme in Computer Science
2025 – 2026
  • Merit-based scholarship for academic excellence and technical project portfolio
  • Coursework: Advanced Networking (B), Research Methodology (B), Decision Support Systems (A)
Integrated Dual Degree Programme (B.Tech + M.Tech)
2021 – 2024
  • Core subjects: Machine Learning (B+), Computer Networks (B+), Software Engineering (B+)
  • Labs: ML with Python (A+), Deep Learning (A+), Computer Networks (A+)

Technical Skills

Programming & Development

Python C++ Bash SQL REST APIs FastAPI gRPC Protobuf

AI & Machine Learning

PyTorch Transformers BERT/GPT RNNs/LSTMs CNNs Federated Learning Explainable AI LiDAR Processing

Cloud & DevOps

Azure Docker Kubernetes Git CI/CD MLOps

Automation & Systems

ML Pipelines Cloud-Native ROS Gazebo Observability Sensor Fusion

Data & Analytics

Pandas NumPy ETL Pipelines Data Compression Data Lakes

Contact

Get in touch for research collaborations, opportunities in ML/AI systems, or to discuss innovative projects.