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

Hi, I'm Deep Shahane, a passionate computer scientist specializing in AI, machine learning, and computer vision. I love building innovative solutions, like optimizing drone-based 3D reconstruction and developing multilingual language translators. My work has ranged from research at the University at Buffalo to hands-on engineering roles in industry. I'm keen on leveraging state-of-the-art technologies like CUDA, PyTorch, and LLaMA to solve real-world problems, especially in AI-powered applications. I'm always eager to learn, collaborate, and contribute to cutting-edge projects that push the boundaries of machine learning and AI.…Hi, I'm Deep Shahane, a passionate computer scientist specializing in AI, machine learning, and computer vision. I love building innovative solutions, like optimizing drone-based 3D reconstruction and developing multilingual language translators. My work has ranged from research at the University at Buffalo to hands-on engineering roles in industry. I'm keen on leveraging state-of-the-art technologies like CUDA, PyTorch, and LLaMA to solve real-world problems, especially in AI-powered applications. I'm always eager to learn, collaborate, and contribute to cutting-edge projects that push the boundaries of machine learning and AI.

Deep Shahane





Hi, I'm Deep Shahane, a passionate computer scientist specializing in AI, machine learning, and computer vision. I love building innovative solutions, like optimizing drone-based 3D reconstruction and developing multilingual language translators. My work has ranged from research at the University at Buffalo to hands-on engineering roles in industry. I'm keen on leveraging state-of-the-art technologies like CUDA, PyTorch, and LLaMA to solve real-world problems, especially in AI-powered applications. I'm always eager to learn, collaborate, and contribute to cutting-edge projects that push the boundaries of machine learning and AI.…Hi, I'm Deep Shahane, a passionate computer scientist specializing in AI, machine learning, and computer vision. I love building innovative solutions, like optimizing drone-based 3D reconstruction and developing multilingual language translators. My work has ranged from research at the University at Buffalo to hands-on engineering roles in industry. I'm keen on leveraging state-of-the-art technologies like CUDA, PyTorch, and LLaMA to solve real-world problems, especially in AI-powered applications. I'm always eager to learn, collaborate, and contribute to cutting-edge projects that push the boundaries of machine learning and AI.

Available to hire

Hi, I’m Deep Shahane, a passionate computer scientist specializing in AI, machine learning, and computer vision. I love building innovative solutions, like optimizing drone-based 3D reconstruction and developing multilingual language translators. My work has ranged from research at the University at Buffalo to hands-on engineering roles in industry.

I’m keen on leveraging state-of-the-art technologies like CUDA, PyTorch, and LLaMA to solve real-world problems, especially in AI-powered applications. I’m always eager to learn, collaborate, and contribute to cutting-edge projects that push the boundaries of machine learning and AI.

Skills

Experience Level

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Language

Javanese

Intermediate

Work Experience

Research Assistant at University at Buffalo [SAIR LAB]

August 1, 2024 - Present

Achieved 16× speedup in drone-based 3D reconstruction by integrating 2D Gaussian Splatting into GPU accelerated SLAM pipelines, outperforming traditional CPU methods. Customized a CUDA optimized differentiable Gaussian surfel rasterizer for 2DGS based SLAM using C++ and CUDA, tuning GPU performance for real-time aerial mapping and benchmarking on TUM RGB-D sequences.

AI Engineer Intern at Bragr

January 31, 2025 - July 20, 2025

Architected and deployed a FastAPI-based chat backend integrated with Azure OpenAI and RAG, reducing response time by 66% and enabling real-time contextual replies with historical data answering sports related queries. Designed an asynchronous web scraping and semantic search pipeline using Python and aiohttp supporting 10+ concurrent users and optimizing MongoDB access. Collaborated with frontend engineers to build a scalable microservices-based chat system, optimizing backend performance and deployment using Azure.

ML Intern at VIT

August 31, 2021 - July 20, 2025

Boosted application relevance and reduced manual effort by launching the VIT Internship Portal with a recommendation system using content-based filtering (TF-IDF and cosine similarity). Built and deployed an end-to-end backend system using Flask and SQLite, integrating scraped Internshala data and a student database of 400 entries to support personalized internship suggestions. Deployed the full-stack solution on Azure using open source services (Q&A Maker, Azure Blob, Azure Maps, Cognitive Speech-to-Text, Azure App Service) to improve accessibility and interaction.