Jun 2019 – Jan 2021 · 1 yr 7 mos

• I lead the Data Science Platform and also the Search team at Embibe.
• I worked closely with the Head of Search at Pinterest/Walmart Labs – Somnath Banerjee.
• https://www.linkedin.com/in/somnath-banerjee/
• Lead Search Team:
• 1. Search Re-Architecture
• 2. Research Failed Queries
• 3. Search V2
• Leading DataScience Platform:
• 1. Houses 50+ DataScience Apps for Showcasing
• 2. Pushed several Data Science APIs to Production
• 3. We support 1M users/second

Apr 2018 – Aug 2018 · 4 mos · Bengaluru Area, India

• The Inkers design and develop deep large-scale AI and ML vision solutions for Enterprises, Defence and Security Organizations.
• The Inkers are working on some of the most complicated and technically challenging problems in the field of AI & Computer Vision, Auto-Navigation, Path Planning, Segmentation, 3D-Reconstruction, Object tracking, Deep Neural Network Optimization, DSP HW integration for Inferencing, Video Compression, etc.
• I did a project on Uber's SBNET (Sparse Block Networks)
• https://eng.uber.com/sbnet/
• On Text Detection and OCR using Pixel-Link and Calamari on Vehicle license Plates and Cheque Amount and Numbers.
• https://github.com/ZJULearning/pixel_link
• https://github.com/Calamari-OCR/calamari
• Successfully cleared all 3 stages of the course!
• https://inkers.ai/

Dec 2017 – Jun 2019 · 1 yr 6 mos · Bengaluru Area, India

• Improving Business Automation using AI/ML and Computer Vision:
• 1. Bank Statement Parser for Financial analysis.
• 2. Cheque Parser using computer vision.
• 3. AI based KYC (Match image from pancard, aadhaar and selfie).
• 4. Classification of images and determine whether an image is a retail shop or not.
• 5. Credit Default Prediction.
• 6. Credit Scorecard.
• 7. Probability of repaying Loan.
• 8. Probability of bouncing a cheque.
• Improving Business Decision/Business Intelligence:
• 1. All analysis based on Monthly, Quarterly, Half-Yearly, Yearly and 3 Years breakdown.
• 2. BI using Google Data Studio.
• 3. Event based triggers to alert internal team, consumers and co-lenders.
• 4. Auto generation of reports and delivery of BI data.
• 5. Realtime Sync from Mongodb to PostgreSQL.
• Microservices Architecture , Scaling and CI/CD:
• 1. Build scalable microservices architecture using AWS, Flask, Zappa(Lambda) and ECS .
• 2. Use all AWS services like Cognito, SNS, Api-Gateway, Secrets Manager.
• 3. Use DB like MongoDB and Postgresql.
• 4. Gitlab as repository manager and CI/CD.
• 5. Complete Automation of Cibil and NACH Payment.
• Data Collection using Client app:
• 1. Built Android Apps for Agents and Retailers.
• 2. Including UI/UX for every screen.

Sep 2015 – Nov 2017 · 2 yrs 2 mos · Bangalore + Bombay

• www.letsgathr.com || www.facebook.com/letsgathr
• Gathr is a mobile and web application that taps into your extended social and place-based networks to find people who want to do what you want to do when you want to do it by matching momentary impulses and general interests. Gathr is like a Meetup for casual, short-notice, social activities. You can tap into your network of friends of friends to watch a movie tomorrow, tap into your neighbourhood network to set up a playdate for your kids over the weekend, or tap into your network of colleagues to see who wants to play a game of foosball in the afternoon.
• Currently we also do,
• High-quality alternative social experiences (for now). We do what we do because we want to create entertainment and more generally, human interactions, that make you feel big, not small. We also work with companies as a Design Agency that reflects the truth of their product through a series of physical interventions.
• Led data for the gathr product from 100 users to 2500 users by doing numerous small experiments.
• Activities: growth analytics, experimentation, product machine-learning, reporting

Jul 2011 – Jul 2015 · 4 yrs

• We built mobile applications that leverage location and social information for social enterprises.

Jun 2007 – Jun 2011 · 4 yrs