I am a Software Engineer who builds scalable backend systems and AI-powered applications.
My work focuses on taking advanced AI/ML concepts and turning them into production-ready systems — including real-time data pipelines, LLM applications, and secure AI services.
- 🎓 B.Tech CSE @ PES University (CGPA: 8.79)
- 🔬 Research Intern @ CCNCS (AI Security & VLM Systems)
- 📄 Accepted & Presented: CrossGuard-VLM (ICCCMLA 2025, Germany)
- 💡 Interests: Backend Engineering, Distributed Systems, AI/ML, Big Data
- Scalable backend systems using FastAPI & REST APIs
- Real-time data pipelines (Kafka, Spark)
- AI-powered applications (LLMs, RAG, NLP, CV)
- Secure systems (input validation, adversarial robustness)
- Data-intensive systems with efficient processing
Languages: Python, C++, Java, C, JavaScript
Backend: FastAPI, Node.js, REST APIs
Databases: MySQL, MongoDB, FAISS, ChromaDB
Big Data: Kafka, Spark, Hadoop
AI/ML: PyTorch, TensorFlow, Hugging Face, Scikit-learn
Tools: Docker, Git, VS Code
| Project | Engineering Focus |
|---|---|
| LineageSentinel | Built a real-time distributed system for detecting data poisoning in streaming pipelines using Kafka & Spark, with automated risk scoring and mitigation |
| PARR-MHQA | Developed a backend LLM application with RAG pipeline, multi-hop reasoning, and hallucination detection for reliable QA systems |
| CrossGuard-VLM | Engineered a modular security system for multimodal AI pipelines to detect and mitigate prompt injection attacks |
| LeakWatch | Built a middleware service for detecting and masking sensitive data across text, images, and audio before AI processing |
| Project | Engineering Contribution |
|---|---|
| Fragment Matching (3D System) | Designed a 3D data processing pipeline for matching fragment point clouds using geometric feature extraction and similarity search |
| FM-TS Interface | Built a modular forecasting framework with pluggable model interfaces and multi-strategy execution (recursive & direct) |
| DQM-ML (Graph System) | Developed a graph-based anomaly detection system with temporal modeling and agent-based decision pipeline |
- Built real-time streaming systems handling complex data pipelines
- Designed modular and extensible architectures for ML systems
- Integrated AI models into production-style workflows
- Developed systems with high performance and reliability
- 📄 Research Paper Accepted & Presented — ICCCMLA 2025 (Germany)
- 🥇 Dr. MRD Scholarship (Top 5%)
- 🥈 Prof. CNR Rao Scholarship (Top 20%)
- 🏅 NTSE District Rank 3