About Me
Technical Skills
Projects
Multi-Agent Backend Code Generation & Reliability Framework
End-to-end system for automatic backend code generation using multi-agent orchestration with reliability guarantees and structured evaluation.
Agentic Multi-RAG Generative AI Framework
Advanced retrieval-augmented generation system with agentic capabilities for intelligent information synthesis and contextual response generation.
MCP-Based LLM Tool Execution Server
Model Context Protocol server enabling secure, scalable tool execution for large language models with function calling capabilities.
Deep Learning Implementations
Comprehensive repository of deep learning architectures including CNNs, RNNs, LSTMs, and transformer models with detailed implementations.
Machine Learning Repository
Collection of machine learning algorithms and techniques from classical methods to advanced ensemble approaches with practical applications.
PyTorch Utilities
Custom PyTorch modules, utilities, and training pipelines for accelerated deep learning experimentation and research.
Optuna Library Implementations
Advanced hyperparameter optimization using Optuna with custom pruners, samplers, and automated ML pipelines.
Managing Imbalanced Data
Techniques and strategies for handling class imbalance in machine learning including SMOTE, undersampling, and cost-sensitive learning.
Experience
AI Intern
- LLM orchestration and multi-agent system design
- Backend AI architecture implementation
- Benchmarking and evaluation frameworks
- System reliability and failure reduction strategies
Research
Multi-Agent Backend Code Generation & Reliability Framework
Published research on developing reliable multi-agent systems for automatic backend code generation. The framework achieves significant improvements in accuracy and reliability through structured agent orchestration and comprehensive evaluation metrics.
View Research Paper