Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am a Computer Vision and Machine Learning Engineer and a PhD candidate in Computer Engineering, with over a decade of industry and research experience in deep learning, biometric recognition, and applied AI. I specialize in building and training models in PyTorch and TensorFlow and delivering end-to-end pipelines across multispectral and multimodal data. I have led production deployments at Samsung R&D Institute Bangladesh, authored several peer-reviewed publications, and currently explore LLMs, retrieval-augmented generation, and agentic-AI systems in academia and independent consulting.…I am a Computer Vision and Machine Learning Engineer and a PhD candidate in Computer Engineering, with over a decade of industry and research experience in deep learning, biometric recognition, and applied AI. I specialize in building and training models in PyTorch and TensorFlow and delivering end-to-end pipelines across multispectral and multimodal data. I have led production deployments at Samsung R&D Institute Bangladesh, authored several peer-reviewed publications, and currently explore LLMs, retrieval-augmented generation, and agentic-AI systems in academia and independent consulting.

Ananya Zabin

Data Scientist, AI Engineer, AI Developer, +2





I am a Computer Vision and Machine Learning Engineer and a PhD candidate in Computer Engineering, with over a decade of industry and research experience in deep learning, biometric recognition, and applied AI. I specialize in building and training models in PyTorch and TensorFlow and delivering end-to-end pipelines across multispectral and multimodal data. I have led production deployments at Samsung R&D Institute Bangladesh, authored several peer-reviewed publications, and currently explore LLMs, retrieval-augmented generation, and agentic-AI systems in academia and independent consulting.…I am a Computer Vision and Machine Learning Engineer and a PhD candidate in Computer Engineering, with over a decade of industry and research experience in deep learning, biometric recognition, and applied AI. I specialize in building and training models in PyTorch and TensorFlow and delivering end-to-end pipelines across multispectral and multimodal data. I have led production deployments at Samsung R&D Institute Bangladesh, authored several peer-reviewed publications, and currently explore LLMs, retrieval-augmented generation, and agentic-AI systems in academia and independent consulting.

Available to hire

I am a Computer Vision and Machine Learning Engineer and a PhD candidate in Computer Engineering, with over a decade of industry and research experience in deep learning, biometric recognition, and applied AI. I specialize in building and training models in PyTorch and TensorFlow and delivering end-to-end pipelines across multispectral and multimodal data.

I have led production deployments at Samsung R&D Institute Bangladesh, authored several peer-reviewed publications, and currently explore LLMs, retrieval-augmented generation, and agentic-AI systems in academia and independent consulting.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

DaVinci Resolve Studio

Intermediate

MySQL

Intermediate

Language

English

Fluent

Work Experience

Applied AI & LLM Engineer at Independent / Continuing Professional Development

January 1, 2024 - Present

Built an autonomous agentic AI system using the Anthropic Claude API and Model Context Protocol to classify incoming email, perform multi-step tool calls, retrieve external context via Tavily API, and generate responses, with a human-in-the-loop fallback for low-confidence cases. Designed a multi-container Retrieval-Augmented Generation (RAG) pipeline using Haystack and Docker Compose, covering document ingestion, chunking, embedding generation, vector storage, and semantic retrieval.

Graduate Research Assistant at MILAB, University of Georgia

January 1, 2021 - November 1, 2024

Led the FaceSensor-ID project for camera sensor origin identification using transfer learning; built deep pipelines for thermal-to-visible face synthesis (CUT, CycleGAN, Pix2Pix) with keypoint-based alignment and identity classification; implemented conditional DCGAN to synthesize diseased iris images to improve recognition on imbalanced datasets; evaluated periocular and binocular authentication across multiple CNN models; studied ocular pathologies’ impact on deep learning ocular recognition; led development of a large-scale multimodal biometric dataset (135k+ images, 13k+ videos) across visible, NIR, and thermal spectra; built automated ingestion, quality-control, annotation, and versioning pipelines; mentored eight researchers.

Graduate Assistant at West Virginia University

January 1, 2018 - December 1, 2020

Developed deep learning models for iris sensor identification and interoperability using transfer learning (AlexNet, GoogLeNet) with image-quality-based data selection; achieved near-100% cross-validation accuracy across five CASIA sensor classes with as few as 300 images per class; applied BRISQUE-based image-quality filtering and SGDM-optimized training to maintain robust classification under limited-data conditions.

Senior Software Engineer / Team Lead at Samsung R&D Institute Bangladesh

January 1, 2014 - July 1, 2015

Sole technical owner of a production bidirectional language engine (C/C++) supporting multilingual Android UI, maintained in production across the Galaxy flagship fleet for 2.5 years; led a sub-team developing Android View Framework components and language-engine integration for the Galaxy Note 4 Edge; drove UI/UX and system performance improvements; developed OpenGL-based graphics for an augmented-reality system that received the Global Samsung Idea Excellence Award; conducted international field testing for the Galaxy S4.

Software Engineer at Samsung R&D Institute Bangladesh

November 1, 2011 - January 1, 2014

Developed Android UI components and language-engine integration for multilingual support; contributed to production readiness for Galaxy Note 4 Edge; enhanced status-bar performance and collaborated with cross-functional teams.

Software Consultant at Southtech Limited

July 1, 2011 - October 1, 2011

Designed relational database architecture for the Ascend Banking System using SQL and PHP; supported a successful CMMI Level-3 audit through rigorous process documentation.