Pradyumn Pathak

Passionate Computer Vision Engineer with 2.5 years of hands-on experience in Computer Vision including Generative-AI. Known for a unique and practical problem-solving approach, consistently applying theoretical knowledge to tackle challenging problem statements. A creative thinker with an insatiable thirst for learning, always pushing the boundaries of innovation. Highly effective in team environments, adept at collaborating with colleagues to achieve common goals. Proficient in end-to-end solution design, capable of translating ideas into successful implementations.

• Helped improve facial recognition system for facial attendance for up to 10k individuals with runtime accuracy of 98%.
• Qualified all the way to Regional Finals in National Level Robotics Competition E‑YiC.

• Complete Machine Learning fundamentals by Stanford university from Coursera.org
• Neural Networks and deep learning from Coursera.org
  

Medical Assistive Robotic System (M.A.R.S.):

Worked in a team of 6 to design a 6 DOF Dexterous Industrial Scale Robotic Mobile Manipulator for the National Robotics Competition E-YiC.

• Worked on robotics perception, designed and deployed object detection and 3D localization solutions using TensorFlow n 2D and 3D point-cloud data.
• Implemented an algorithm for calculating the optimal grasping position for the object based on its geometrical shape.

Image-Processing-based Rubik's Cube Solving Robot:Worked in a team of 3 to design a vision-based 6 DOF Rubik's cube-solving robot.

• Worked on color classification model using TensorFlow and OpenCV