I'm a Software Engineer passionate about AI, ML, and Backend Development. Currently pursuing my Master's in Software Engineering at Carnegie Mellon University, I enjoy tackling challenging problems at the intersection of AI and production systems.
- 🔭 Currently working on: Full-stack machine learning applications and product information management systems.
- 🌱 Learning: Go, microservices, distributed systems.
- 💼 Seeking: Full-time opportunities in machine learning, AI, and software engineering roles.
I recently delivered a series of lectures on deep learning concepts at Carnegie Mellon University, which are available on YouTube. Check them out below:
- Lecture 1: Introduction to Deep Learning: An overview of deep learning fundamentals, covering neural networks, activation functions, and backpropagation.
- Lecture 2: Debugging Deep Learning Networks: A deep dive into how debugging is performed in deep learning nerual networks.
- Lecture 3: Introduction to Dataloaders: An introduction to dataloaders and how to use them using PyTorch library.
- Lecture 4: Convolution Neural Networks: Introduction to CNNs and its basic building blocks.
- 🔍 Product Information Management System (PIMS): Developed a comprehensive product management solution with ASP.NET, Vue.js, and Azure SQL.
- ⚙️ Distributed Bitcoin Miner: Built a scalable, fault-tolerant distributed system using Go and UDP for efficient job distribution and mining task coordination with concurrency.
- 🧠 Face Classification and Verification: Built CNN-based ResNet-34 models for face classification and verification, ranking in the top 5% on Kaggle.
- 🎤 Speech Recognition System: Achieved 86% accuracy in speech classification using neural networks and frame-level classification.
- 📈 Stock Predictive Model: Developed a predictive model utilizing CNN, LSTM, and Transformer architectures to integrate sentiment analysis with stock price data, achieving a 95% reduction in MSE to 0.00008 compared to initial baseline models.
- 🛠️ Dynamic Memory Allocator: Designed and implemented a dynamic memory allocator in C++ that mimics the standard malloc, free, realloc, and calloc functions, achieving a space utilization of 74% and throughput of 8000 KOPS using segregated free lists for efficient memory management.
Programming Languages:
Frameworks & Libraries:
Databases:
Cloud & DevOps:
Version Control & Collaboration:
