Narayanan Elavathur Ranganatha: Machine Learning (ML) Engineer & CS Graduate
Narayanan Elavathur Ranganatha: Machine Learning (ML) Engineer & CS Graduate
Narayanan Elavathur Ranganatha: Machine Learning (ML) Engineer & CS Graduate
Skilled engineering professional with expertise in deploying Machine Learning methodologies & deep learning frameworks. I have
extensive experience in Reinforcement Learning, Computer Vision, Natural Language Processing and Robotics using Python,
C/C++ with Linux. I am also a continual learner and have completed several MOOCs in Deep Learning and Machine Learning
targeting autonomous system production.
KEY SKILLS
COURSES
WORK EXPERIENCE
Robotics Research Center, IIIT Hyderabad | Research Intern Manipal | May ’19 - Present
● Currently working on building a simulation in Gazebo to replicate human movement in a warehouse.
● The simulation uses data from a Kinect sensor and will be used to train a deep learning model for Human Trajectory
Prediction.
Rammer.ai | Data Science Intern Pune | Dec ‘18 - Jan ‘19
● Extracted ‘action items’ and filtered out chit-chat from speech to text data from meetings.
● Trained Bidirectional GRU with attention mechanism for punctuation restoration model in tensorflow..
● Designed and trained a 2 layers GRU model in Keras to classify an input sentence as relevant after the meeting or not.
● Augmented the training data by analyzing constituency tree and part-of-speech tags, synonym insertion and using Google
Sentence Encoder to extract sentences with the same semantic meaning by using similarity metrics such as cosine similarity.
Project MANAS | Artificial Intelligence Division Member Manipal | Jan ’18 - Dec ‘18
● Deployed reinforcement algorithms like actor-critic with experience replay, Q-Learning & Behavioral Cloning for testing the
same with real world automobile simulators such as Airsim.
● Replicated reinforcement learning algorithms such as Generative Adversarial Imitation Learning, actor-critic with experience
replay and tested them in various MuJoCo environments.
● Worked on video compression using contextual recurrent neural networks.
KEY PROJECTS
Road Segmentation using satellite images Oct ‘18 - Nov ‘18
● Trained a Dilated U-net for segmentation of roads in satellite images collected via the MapBox API.
Behaviour Cloning. Oct ‘18 - Nov ‘18
● Trained a convolution neural network by collecting driving data on a simulator to teach the car to track lane centers.
Vehicle Tracking and Detection May ‘18 - Jun ‘18
● Trained a Support Vector Machine (SVM) using Histogram of Oriented Gradients (HOG) to identify car and non-car images.
● Used windows of different sizes across frames of video (smaller windows for cars far away and larger windows for closer
ones)
Reinforcement Learning Dec ‘17 - Feb ‘18
● Applied Deep Q-Learning, Actor-Critic & Proximal Policy Optimization in OpenAI gym environments to perform continuous
and discrete control tasks.
Artificial Intelligence Subsystem Member | DRONAID A pr ’17 - Aug ’17
● Implemented ROS concepts to extract 3-D mapping data from a camera & enable data transfer between the camera & GPU.
Avionics Subsystem Member | Thrust MIT Mar ’17 - Apr ‘17
● Applied Arduino UNO to extract bitwise data & implemented scaling to convert hexadecimal values into engineering units.
● Deployed filters like low pass, root mean square, etc. to condition signals & extract accurate information from the sensors.
EDUCATION
B.E. – Computer Science Engineering | Manipal Institute of Technology| CGPA-9.11 Manipal | Aug ‘16 - Present
All India Senior School Certication Examination | Apeejay School NOIDA | 92.80% NOIDA | Mar ‘16