Developed a Classifier to classify an email as Receipt and Non Receipt.
- Used Python Scikit learn
- For extracting the data we used NLP library(Spacy)
- Custom trained the NLP model using NER (Named Entity Recognition)
Process Automation System
- Have developed this platform using MERN tech stack.
- Integrated payment gateways and notification alert system
- Improved OEE (Overall Equipment Effectiveness) by 11%.
Worked on a Digital Signage platform for a fortune 500 company and solely handled the widget management system. Integrated APIs from various platforms like Twitter, Facebook, Instagram, Google Calendar, Outlook, Yahoo News etc and developed a platform to maintain real time synchronization.
- Used celery to maintain queues for making external API calls, made sure that the number of API calls has a threshold and is called only permitted number of times (to avoid spamming).
- Postgres was used to maintain data and periodic tasks were allocated to clear prune/unused data.
- Database backup is stored in s3 bucket.
Worked on Android App built from Scratch.
1. Optimized in terms of battery usage and network calls.
2. Reduced APK size by applying Image compression and using ProGuard.
3. Published apps in different flavors (Paid/free and based on country)
4. Implemented native language support (with RTL and LTR for supporting Farsi language)
5. Offline app with SQL Support and data caching (Used Android Architectural Components like Room Db)
6. Implemented Android Widgets.
- Worked on multiple projects which includes Educational platform, Food Delivery and Kitchen Management platform, Online Coupon Management platform etc.
- Exploring Machine Learning and its applications in the real world.