I'm a Senior Data Scientist with 3 years of industry experience.
My mission is to inspire enthusiasts to break in and then grow in the world of data science! Currently, I work at OCTAVE - John Keells Holdings, where I solve impactful business problems for the Retail sector by unlocking the power of data.
Before joining OCTAVE, I worked at Trabeya - An AI startup, where I was designing, developing, and deploying end-to-end AI products for business problems in the domain of Smart Cities, Data Privacy, Genomics, Aquaculture, and more.
Additionally, I write articles on Data Science & Machine Learning, build pet-projects, invest time in continual development through MOOCs, and make a conscious effort to share the knowledge.
Summing it all up, I learn, create and share.
You'd see me working, writing and mentoring on ML/AI most of the time. I'm fairly comfortable with PyTorch and TensorFlow. I've worked on projects using both the frameworks.
Python is my go-to language when it comes to coding up something. In the past I've worked on R, Java and C++ so I find it easy to use these languages as well.
I generally use MySQL databases when it comes to structured data and MongoDB for unstructured data. I've used Gremlin for graph data.
I'm comfortable with Google Cloud Platform and Micorsoft Azure, since I use them mainly for my daily tasks. I've used AWS for some past projects.
I believe every developer needs an understanding of Devops side of things. I package my code in to Docker containers and can configure it into Kubernetes clusters.
A Google CodeU Summer 2019 Project. Features include Hate-speech filters, Automatic generation of hashtags and Text to Speech Generation for user posts.
A project in collaboration with Omdena and RedDot Foundation to collect data from various sources and model them to understand the state of Domestic Violence during COVID-19.
[WRAPPING UP] Conducted Experimental Design and Analysis, built predictive models leveraging the results and Proposed an optimal promotional strategy for Starbucks.
Classification of day-night images with an accuracy of 93.75% through Image Processing using OpenCV and zero Machine Learning.
A Secure and Private AI 2019 Project. Prototype of Rescue at Sea mission through detection and tracking of people using OpenCV and PyTorch.