Developed a car rental web application that suggests optimized trips based on user preferences. Integrated chatbase for intelligent recommendations, implemented secure authentication, and connected multiple APIs to provide a seamless booking experience.
"}},{"node":{"frontmatter":{"title":"Sexism Detection with BiLSTM","cover":{"childImageSharp":{"gatsbyImageData":{"layout":"constrained","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/79b805fe7c57375025237b7f0bb4ea98/68012/demo.png","srcSet":"/static/79b805fe7c57375025237b7f0bb4ea98/17f1f/demo.png 175w,\n/static/79b805fe7c57375025237b7f0bb4ea98/527dd/demo.png 350w,\n/static/79b805fe7c57375025237b7f0bb4ea98/68012/demo.png 700w,\n/static/79b805fe7c57375025237b7f0bb4ea98/70b84/demo.png 1400w","sizes":"(min-width: 700px) 700px, 100vw"},"sources":[{"srcSet":"/static/79b805fe7c57375025237b7f0bb4ea98/78482/demo.avif 175w,\n/static/79b805fe7c57375025237b7f0bb4ea98/d637e/demo.avif 350w,\n/static/79b805fe7c57375025237b7f0bb4ea98/5acd9/demo.avif 700w,\n/static/79b805fe7c57375025237b7f0bb4ea98/e0697/demo.avif 1400w","type":"image/avif","sizes":"(min-width: 700px) 700px, 100vw"},{"srcSet":"/static/79b805fe7c57375025237b7f0bb4ea98/98832/demo.webp 175w,\n/static/79b805fe7c57375025237b7f0bb4ea98/4a3fb/demo.webp 350w,\n/static/79b805fe7c57375025237b7f0bb4ea98/8712a/demo.webp 700w,\n/static/79b805fe7c57375025237b7f0bb4ea98/0129f/demo.webp 1400w","type":"image/webp","sizes":"(min-width: 700px) 700px, 100vw"}]},"width":700,"height":429}}},"tech":["Python","Deep Learning","BiLSTM","NLP","GloVe","TensorFlow"],"github":"","external":"","cta":null},"html":"Built a Bidirectional LSTM model to classify and categorize online sexist content using the SemEval 2023 EDOS dataset. Preprocessed imbalanced real-world text data and implemented GloVe embeddings for binary classification, achieving robust performance on detecting harmful content.
"}},{"node":{"frontmatter":{"title":"BanglaMUX: Regional Dialect Detection & Translation","cover":{"childImageSharp":{"gatsbyImageData":{"layout":"constrained","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/79b805fe7c57375025237b7f0bb4ea98/68012/demo.png","srcSet":"/static/79b805fe7c57375025237b7f0bb4ea98/17f1f/demo.png 175w,\n/static/79b805fe7c57375025237b7f0bb4ea98/527dd/demo.png 350w,\n/static/79b805fe7c57375025237b7f0bb4ea98/68012/demo.png 700w,\n/static/79b805fe7c57375025237b7f0bb4ea98/70b84/demo.png 1400w","sizes":"(min-width: 700px) 700px, 100vw"},"sources":[{"srcSet":"/static/79b805fe7c57375025237b7f0bb4ea98/78482/demo.avif 175w,\n/static/79b805fe7c57375025237b7f0bb4ea98/d637e/demo.avif 350w,\n/static/79b805fe7c57375025237b7f0bb4ea98/5acd9/demo.avif 700w,\n/static/79b805fe7c57375025237b7f0bb4ea98/e0697/demo.avif 1400w","type":"image/avif","sizes":"(min-width: 700px) 700px, 100vw"},{"srcSet":"/static/79b805fe7c57375025237b7f0bb4ea98/98832/demo.webp 175w,\n/static/79b805fe7c57375025237b7f0bb4ea98/4a3fb/demo.webp 350w,\n/static/79b805fe7c57375025237b7f0bb4ea98/8712a/demo.webp 700w,\n/static/79b805fe7c57375025237b7f0bb4ea98/0129f/demo.webp 1400w","type":"image/webp","sizes":"(min-width: 700px) 700px, 100vw"}]},"width":700,"height":429}}},"tech":["Python","Transformer Models","Whisper","BERT","Bangla T5","NLP"],"github":"","external":"","cta":null},"html":"Undergraduate thesis project focused on developing a transcription and translation model for Bangla dialects using transformer-based approaches. Built a pipeline that takes regional dialects of Bangla language and converts them into textual standard Bangla. Explored various tools and models including Whisper, BERT, Bangla T5, and mT5 for low-resource language processing.
"}},{"node":{"frontmatter":{"title":"DeshScan — Private On-Device Document Scanner","cover":{"childImageSharp":{"gatsbyImageData":{"layout":"constrained","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/909a848ba7c44d6c469e2c39ed3ff2ea/a9ed7/demo.jpg","srcSet":"/static/909a848ba7c44d6c469e2c39ed3ff2ea/488be/demo.jpg 175w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/6dc06/demo.jpg 350w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/a9ed7/demo.jpg 700w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/c0c9b/demo.jpg 1400w","sizes":"(min-width: 700px) 700px, 100vw"},"sources":[{"srcSet":"/static/909a848ba7c44d6c469e2c39ed3ff2ea/b40e8/demo.avif 175w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/c9a2b/demo.avif 350w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/5f582/demo.avif 700w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/37afc/demo.avif 1400w","type":"image/avif","sizes":"(min-width: 700px) 700px, 100vw"},{"srcSet":"/static/909a848ba7c44d6c469e2c39ed3ff2ea/c5506/demo.webp 175w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/6ac18/demo.webp 350w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/22dbc/demo.webp 700w,\n/static/909a848ba7c44d6c469e2c39ed3ff2ea/d3f82/demo.webp 1400w","type":"image/webp","sizes":"(min-width: 700px) 700px, 100vw"}]},"width":700,"height":272}}},"tech":["Python","PyTorch","U-Net","OpenCV","OCR"],"github":"https://github.com/munawaransary/deshscan","external":"","cta":null},"html":"A free, private document scanner where your documents never leave your device. Photograph any document — NID card, certificate, form — and get a clean, flattened, shareable scan. A compact U-Net, trained on self-generated synthetic shadow pairs, removes shadows and color casts, and the whole pipeline (detect → dewarp → shadow removal → OCR → PDF) runs locally, even on CPU. Supports Bangla + English OCR.
"}},{"node":{"frontmatter":{"title":"MonCare — Trilingual Mental-Health Support Chatbot","cover":{"childImageSharp":{"gatsbyImageData":{"layout":"constrained","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/48d50fbdefac8970002586bcb13d212e/76f9a/demo.png","srcSet":"/static/48d50fbdefac8970002586bcb13d212e/9051f/demo.png 175w,\n/static/48d50fbdefac8970002586bcb13d212e/b2c0b/demo.png 350w,\n/static/48d50fbdefac8970002586bcb13d212e/76f9a/demo.png 700w,\n/static/48d50fbdefac8970002586bcb13d212e/36422/demo.png 1400w","sizes":"(min-width: 700px) 700px, 100vw"},"sources":[{"srcSet":"/static/48d50fbdefac8970002586bcb13d212e/2f9bd/demo.avif 175w,\n/static/48d50fbdefac8970002586bcb13d212e/56394/demo.avif 350w,\n/static/48d50fbdefac8970002586bcb13d212e/565bd/demo.avif 700w,\n/static/48d50fbdefac8970002586bcb13d212e/c4e48/demo.avif 1400w","type":"image/avif","sizes":"(min-width: 700px) 700px, 100vw"},{"srcSet":"/static/48d50fbdefac8970002586bcb13d212e/9b040/demo.webp 175w,\n/static/48d50fbdefac8970002586bcb13d212e/18cc4/demo.webp 350w,\n/static/48d50fbdefac8970002586bcb13d212e/efdb4/demo.webp 700w,\n/static/48d50fbdefac8970002586bcb13d212e/1f42c/demo.webp 1400w","type":"image/webp","sizes":"(min-width: 700px) 700px, 100vw"}]},"width":700,"height":372}}},"tech":["Python","RAG","FAISS","GPT-OSS-20B","FastAPI"],"github":"https://github.com/munawaransary/moncare","external":"","cta":null},"html":"A private, trilingual (Bangla · Banglish · English) mental-health support chatbot that runs entirely on one local GPU. It detects the user's language, screens every message through a two-stage crisis gate, and answers with retrieval-augmented generation over a 19K-item counseling knowledge base (bge-m3 embeddings, FAISS, cross-encoder rerank, local LLM). Crisis responses are deterministic and route to verified human helplines, with an output guard that strips any helpline number the model invents. Built as a Research Engineer Intern at UIU's AIMS Lab. Not a medical device.
"}}]}}}