Akhil Santha

Contact: Fremont, San Jose, CA, ****8 585-***-**** adegtw@r.postjobfree.com LinkedIn: https://www.linkedin.com/in/akhilsantha/ SUMMARY

• Deep Learning Researcher with 2+ years of experience in building and implementing deep neural networks for various complex domain tasks. Passionate engineer and thriving researcher with the ability to apply machine learning and deep learning techniques and algorithm development to solve real world industry problems. I am open to relocation anywhere within the US.

PROFESSIONAL EXPERIENCE

Deep Learning Research Intern, KODAK ALARIS, ROCHESTER, NY Feb 2019 – Dec 2019

• Facial understanding: Built deep learning models for face localization, detection and recognition and eye glass detection. Developed algorithms for deploying the trained models in the Kiosks using OpenCV dnn libraries. Achieved 99% accuracy on eye glass detection. Our models surpassed the Microsoft Azure’s performance by 5%.

• Deep Learning for 3D data: Classification and Segmentation of 3d point cloud data using deep learning models. Built convex hulls for 3d point cloud data using point cloud libraries.

• Technologies used: TensorFlow, OpenCV, Dlib, C++. Research Assistant, VISUALDX, ROCHESTER, NY Aug 2018 – Jan 2019

• Deep Learning Research: Built Deep learning models for the classification of skin diseases. Used computer vision preprocessing techniques on dataset for boosting the performance. Achieved 61% accuracy on SD198 dataset which is 3% more than state-of-the-art performance.

• Technologies used: TensorFlow, Keras, OpenCV

Deep Learning Research Assistant, MACHINE INTELLIGENCE LAB, RIT May 2018 – May 2020

• Thesis: Generation of Deepfakes using LSTM based GANs via self-supervised training.

• Sign Language Recognition: Building multiple deep learning models for sign language recognition. We have achieved state- of-art BLEU scores on American, Chinese and German continuous sign language datasets.

• Technologies: Pytorch, TensorFlow.

EDUCATION

ROCHESTER INSTITUTE OF TECHNOLOGY, ROCHESTER, NY

• Master of Science, Jan 2018 - May 2020, GPA: 3.41/4.0 UNIVERSITY COLLEGE OF ENGINEERING, OSMANIA UNIVERSITY, HYDERABAD, INDIA

• Bachler of Science, Aug 2013 - May 2017, GPA: 4.0/4.0 SKILLS

PROGRAMMING LANGUAGES & FRAMEWORKS: Python, C, C++, MATLAB/OCTAVE, TensorFlow, Keras, PyTorch, Caffe PACKAGES: Scikit-Learn, Pandas, Matplotlib, NumPy, SciPy, NLTK. RELEVENT COURSES: Machine learning, Deep learning, Computer Vision, Brain Inspired Computing, Parallel Computing, Embedded Systems, AWS-Machine Learning, Amazon SageMaker. PROJECTS

SKIN DISEASE CLASSIFICATION: Built deep neural networks to classify skin diseases using Pytorch framework. 3D POINT CLOUD CLASSIFICATION: Classification and Segmentation of 3d data using deep learning models in TensorFlow framework TRAFFIC SIGN CLASSIFICATION: German traffic sign classification using CNNs in Keras framework. PARALLEL COMPUTING: Implemented Hodgkin Huxley Neuron model and ray tracer on multi-core chips and analyzed speedups using openMPI.

ACTIVITY RECOGNITION: Developed graph convolutional neural networks for sign language recognition using Pytorch framework. We used openpose for extracting the body points from the datasets.