Dipanshu is highly skilled engineer with a strong foundation in Python, specializing in data processing and transformation using Pandas and NumPy. Dipanshu displays proficiency in SQL, machine learning, and cloud technologies (AWS). He is experienced in building efficient data pipelines and optimizing query performance. Dipanshu is also committed to writing clean, maintainable, and scalable code.
Creating and developing innovative software solutions for clients across a broad range of industries.
Participating in scrums consisting of cross-functional teams, both software and hardware.
Ensuring that features are being delivered efficiently and on time.
Working on multiple software solutions focused on solving the problem of document-based information overload for the financial services industry.
Leading new projects using NLP, Machine Learning, data mining (unstructured data from PDF documents) + scraping (intelligent goal-oriented, config-based), asynchronous APIs, microservices, GNU/Linux (RHEL, Ubuntu), and on-premise deployments.
Enhancing the functionalities of current software systems and creating predictive models for ML-based features.
Worked on a Python/Django-based back end for an online marketplace platform that connects brands with social media influencers.
Implemented performance tuning and automation for the application.
Developed back-end components to enhance performance and receptiveness, server-side logic, and statistical learning models for the application.
Worked on the development of an online marketplace platform for vehicle servicing/repairs on demand at any location.
Utilized LAMP stack, hosted and managed with AWS.
Worked on the complete software development process from conception to deployment.
Extracts transaction data from any bank statement PDF and generates a financial health model for credit rating agencies.
Developed internal python libraries for PDF-related data mining and a flexible, robust PDF processing library better than open-source options.
Participated in integration with a client's existing systems via on-premise deployment
AI-powered bot sends custom telegram messages containing affirmations to users.
Users can manage topics, add/remove them at any time, and request new affirmations outside the schedule.
The goal of the bot is to help users change their self-perceptions and improve themselves daily.
Developed large parts of the KEngine back-end and the self-learning mechanisms.
Implemented user-led training mechanism for KEngine to extract key & value pairs from PDF documents.
Utilized technologies like Python, PDF processing libraries, HTML+CSS+JS, OCR tools, SpaCy, NLTK in the KEngine project.
Developed News Analytics that sources financial news and adds tags to specific companies
Classified news into categories and predicted user preferences using Python, NLTK, SpaCy, TensorFlow, Selenium, BS4, Flask
Lead Developer for the entire 'News Analytics' operation from news sourcing to user preference prediction
FinView is a SaaS tool designed for auto-generating hyper-personalized videos.
The tool primarily focuses on the finance industry, converting data from reports and PDFs into videos.
The software has the ability to create videos at a large scale, making it suitable for managing an entire customer base.
Developed an AI-based web app Clickable.boo for YouTube creators
The app provides instant detailed feedback on thumbnails and titles before publishing
This tool helps to maximize chances of high view counts
