Integrated MTech in Software Engineering
Vellore Institute of Technology, Vellore
CGPA: 8.01 / 10
Generative AI · Full-stack · Machine Learning
NAVEEN RAJ M
Final Year Software Engineering Student
Profile
Professional summary
Passionate software engineering student with hands-on experience in full-stack web development, data analytics, and machine learning. Proficient in building secure, scalable web applications from robust backend APIs and database design to responsive frontend interfaces. Experienced in extracting insights from data and integrating AI and ML capabilities into real-world workflows. Grounded in core computer science principles with a practical, build-first approach. Eager to work on challenging problems alongside teams that value engineering quality, intelligent systems, and continuous learning.
Background
Education
Higher Secondary Education (Class XII)
Sunbeam Matric Higher Secondary School, Vellore
Percentage: 84.1%
Selected work
Projects
OneSpot Finance
Personal Finance & Asset Management Platform
Node.jsExpressPostgreSQLReactREST API
- Engineered a full-stack personal finance platform integrating expense tracking, investment management, liabilities, financial goals, and dashboard analytics into a unified application.
- Deployed the application on Render with a live database and real-time stock price updates, while continuously expanding the platform with additional financial modules and features
Flower Classification
Convolutional Neural Networks
PythonTensorFlowCNNKerasPillowNumPy
- Developed a multi-class flower image classification system using Convolutional Neural Networks (CNN) with Conv2D, MaxPooling, ReLU, and Softmax layers to classify 5 flower categories.
- Implemented a health analysis module using RGB-to-HSV color analysis to estimate visible dryness, overwatering, and poor lighting conditions from flower images using Pillow and NumPy.
Real-Time RAG-Based Hallucination Firewall for LLMs
Ongoing SDD Project
PythonRAGOllamaChromaDBFAISSNLI
- Designed a real-time hallucination verification architecture for LLM-based banking customer support to reduce unsupported or contradictory information reaching end users.
- Proposed a claim-level verification pipeline combining RAG, atomic factual claim extraction, trusted evidence retrieval, and Natural Language Inference (NLI) for real-time response validation.
- Designed an intervention workflow to classify generated claims as entailment, contradiction, or unsupported, followed by suppression or evidence-grounded regeneration before final response delivery.
Toolbox
Technical skills
Core CS & Programming
- Java
- Data Structures
- Computer Networks
- OOP
- DBMS
- Operating Systems
Authentication & Security
- JWT Authentication
- Password Hashing (bcrypt)
Backend & Databases
- Node.js
- Express.js
- REST APIs
- PostgreSQL
AI & Machine Learning
- Model Training & Optimization
- Deep Learning
- LLM and RAG Fundamentals
Methodologies & Tools
- SDLC
- Agile (Scrum)
- GitHub
- Postman
Credentials
Certifications
Full Stack Web Development
Udemy
Generative AI Leader
Beyond code
Interests
Technical
- Generative AI
- Cloud Technologies
- Machine Learning
Non-technical
- Videography
- Editing
- Stock Market Analysis
Let's talk
Contact
Open to internships, collaborations, and teams that care about solid engineering and continuous learning. Prefer email or LinkedIn for first contact.