Hello! I'm Juan Camacho Mohedano, an AI researcher and robotics enthusiast passionate about pushing the boundaries of intelligent systems. 🚀
Currently pursuing my MSc in AI Systems at the University of Trento, I'm exploring cutting-edge research in AI Agents, Large Language Models (LLMs), and Vision-Language Models (VLMs). My journey in AI has taken me from building CNN-based classifiers to implementing advanced test-time adaptation techniques for vision models.
With experience at NEURA Robotics in Germany and a background in robotics competitions (RoboCup), I blend theoretical knowledge with practical applications. My work spans from 6DoF robot grasping to customizing RAG applications for optimized vector search retrieval.
I'm always excited to collaborate on projects that leverage AI to solve real-world problems, particularly at the intersection of computer vision, natural language processing, and robotics.
Projects
- Alpha-CLIP: A Breakthrough in Region-Focused AI Vision - A literature review exploring Alpha-CLIP's capabilities in region-focused AI vision and proposing enhancements for improved performance.
- Test-Time Adaptation for Vision-Language Models - Analysis and improvement of efficient adaptation techniques for vision models facing distribution shifts. Benchmarked failure cases and implemented waiting list mechanism to improve performance on non-IID data streams.
- 6DoF Robot Grasping with TIAGo - My BSc thesis project on teaching robots to grasp objects using deep learning and computer vision. Developed for the LASR team and RoboCup competitions.
Contact
Email: jucamohedano@gmail.com