Welcome to my GitHub! I'm a Machine Learning Engineer with ~7 years of experience building and scaling ML systems across NLP, Computer Vision, and traditional time-series domains. I'm passionate about solving real-world problems using data, models, and elegant engineering and I'm especially drawn to areas at the intersection of research and product.

Currently following interesting research from OpenAI, Meta AI, FAIR, and DeepMind, with a focus on contributing to next-generation models, applied ML, and responsible AI systems.

• Efficient deep learning architectures (Transformers, MobileNets, Distilled Models)
• Multimodel vison-language-sensing for accurate 3D motion prediction for AR/VR/XR
• NLP: Text summarization, Q&A systems, and retrieval-augmented generation (RAG)
• Vision: Visual Question Answering (VQA), self-supervised learning in CV
• ML for personalization, recsys, and user behavior modeling

• object-detection-iOS-app A real-time object detection application for iOS devices that utilizes computer vision to identify and classify objects through the device's camera.

• efficient-cv-models
  Implemented and benchmarked light-weight models like MobileNetV3, SqueezeNet, and EfficientNet-lite on edge devices.

• transformer-from-scratch A minimal yet complete implementation of the Transformer model (Vaswani et al., 2017). Includes positional encoding, multi-head attention, and masking for decoder logic.

• abstractive-summarizer
  Built using Hugging Face + Seq2Seq models (T5, BART). Added evaluation metrics like ROUGE, BLEU, and support for custom datasets.

Python • PyTorch • NLP • 'CV' •ONNX •Transformer•'LLM' •OpenCV•scikit-learn` •'XGBOOST'

• 💼 LinkedIn
• 📫 Email
• 🧠 Medium/Blog

I'm actively looking to contribute to high-impact ML projects and collaborate with teams working on intelligent systems at scale.

Thanks for stopping by! 🌱