Work Experience

Working with a start-up based in Denmark, Project involves researching and developing deep learning and AI based crypto trading models.
Pattern recognition of interesting regions in rolling windows based data format, interesting regions are breakouts of price, takeoff of price etc. building various classification models to predict a region (breakout/takeoff) of a given window and put a trade accordingly.

Building Deep Learning models for forex market prediction, Buy-Sell decision model using python deep learning algorithms.
Financial Indicator based pattern recognition to recognize events based on financial indicator and using the prediction for trading.
Deploying an end to end pipeline which takes data from Kafka server and predict outcomes for near real time algorithmic trading.
Deploying machine learning/deep learning models in python trading algorithm pipeline.
Maintaining windows GPU workstation.

Building Image Classification algorithm for mobile app using Keras and TensorFlow, the use case is to build an algorithm that can classify pictures taken by user from mobile app to ease the search process on lockthedeal.com mobile app. Technique used to build models are transfer learning with retraining last layer for 1000 classes.
Deploying model on Flask-restful API which can be accessed by mobile app.
Maintaining in-house Ubuntu GPU Server for training Deep Learning models.

Bosch connected industries 4.0 Project, where data is collected from sensors installed in various manufacturing units and analysed for trend detection, change point detection, frequency detection for predictive maintenance. Building Flask REST API microservices backend for each of the above mentioned module in python.
Time series Anomaly Detection algorithm in R for sales forecasting of Chinese business unit, replacing anomaly with the average value by month and year.

Developing analytics solutions for a client in retail domain, reporting relevant business questions, such as analysing flow of different categories of Items and modelling discount promotions.
Other Project was Customer Segmentation and understanding Customer Behaviour. Segmenting Customers using Clustering Algorithm (K-means, Hierarchical). Then reporting similarity features for each group or segment such as frequently bought items, location etc.