Hi, I'm Sagar Sanjeev
A
Self-driven, quick starter, passionate programmer with a curious mind who enjoys solving a complex and challenging real-world problems.
About
Experienced Full Stack Developer with a strong experience in Python and Angular/React and AI/NLP. Delivered NLP/Deep Learning-based solutions for a Large North American Bank for their financial and sales prospecting use cases with Accenture Innovation Labs. Holding a US patent in Enterprise Based Nature Langauge Processing. Passionate about developing complex applications that solve real-world problems.
- Languages: Python, JavaScript, TypeScript, C, C++, HTML/CSS, Bash, Shell
- Databases: MySQL, MS-SQL, Neo4j
- Libraries: Pandas, NumPy, OpenCV
- Frameworks: Flask, React, Angular, Node.js, , Material UI
- Tools & Technologies: IOT, Git, Docker, AWS, GCP, Heroku, Jenkins, Quartz
Experience
- Working on a Enterprise Application that provides a LEGO based experience to the users and helps them creating and modifying the data pipelines for their application as per the Standards
- Tech Stack: React, TypeScript, JavaScript, MS-SQL, Python, NodeJS, PCF, Kubernetes
- Worked as Full Stack Developer in Data Governance in Risk Team, building and maintaining an enterprise application for client, that scans, manages and handles metadata and data lineage of many LOB's of the bank
- Developed a System for Translation of Technical Data Transformations into plain English Language using NLP Concepts that saved around 350 person hours and translated around 250K Transformations.
- Managed Fuzzy Search Tool that was built around NLP grammar construct for querying databases using natural language without having to construct an SQL query and filters data from around 23 Million Records
- Managed the automation of the regulatory [FFIEC (4 Reports) and 10Q (3 Schedules)] reports and linking of the same to Metadata using NLP for Regulatory Exams
- Metadata Visualization using Python libraries like networkx and JavaScript libraries like D3.js
- Managed the implementation of Graph Algorithms to find the Centrality, Anomalies and Modularity of the Graph
- Applied NLP to articles scraped from the internet to extract meaningful information using Open information extraction to develop Artificial Intelligence based Search System for Adverse Media Checks (Knowledge Graph)
- Developed and Managed the Queue Based Reporting Module of the Application Independently
- Selected as the Member for Quick POC Development for Client Demos in Accenture
- Conducted training sessions on Python/ UNIX.
- Tech Stack: Python, Angular, Flask, Neo4j, RDMS, NLP, Quartz, Jenkins
Roles and Responsibilities
-
Topic:Discrete Event Simulation of Computer Networks
The Main Outcome of this project is Simulation of Networks/Topologies in Virtual Enviroment before practical implementaion, So that the problems can found out and be solved in a timely manner It also simplifies the development using new technologies that facilitate the creation of a network protocols etc. - Tech Stack: Python, Numpy, Scipy
- Managed the Server Side of Lab including the Database, Implemented Load Balancing and Clustering on the Servers
- Created IOT POC's and Implemented Data Analytics over the data collected from the sensors
- Tech Stack: IOT , AD-RMS, Server Side Management, Clustering, Load Balancing
- At Idea Cellular Limited, The course of work was to understand the newly implemented service in the branch-office of Active Directory Rights Management Service (AD-RMS) and then make then train the staff about the different features and demonstrate it to them so that they use it and keep the digital documents encryted inside the company's network
- Tech Stack: Active Directory Rights Management Service (AD-RMS)
Projects
Extracting Information from Documents and Assign Accordingly
- Extracting Information from Multiple Types of Documents, News, and OCR, then extracting relevant information from them and classifying the document using Natural Language Processing & Deep Learning and by using unique algorithm, combination of a traversal technique to put the extracted information into perspective.
- Tech Stack: NLP Open-Source Libraries & Toolkit, D3, Flask, Python, SQL, AWS
An Application that Scans Documents and Extracts Action Items
- Extracting information using Natural Language Processing & Deep Learning from the different regulations e.g. (GDPR, MIFID2 and so on) and then extracting the context, and then assignment of the directive to the proper dept. of the organization
- Tech Stack: NLP Open-Source Libraries & Toolkit, D3, Flask, Python, GCP, SQL
AI Based Adverse Media Checks for Onboarding new Clients
- Tools: HTML, CSS, Bootstrap, Flask, SQLAlchemy, Postgresql, Python
- Users can view posts and contact the admin via Contact Page.
- Admin can Add, Delete, Update posts.
An POC for Anti-Skimming Protection in ATM's.
- Created a POC system that allows to withdraw money from ATM after confirming Account Holders Identity.
- Technologies Used: OpenCV, Python
Automation in Automobiles using Internet of Things.
- The project included the concept of getting advanced features in Medium Range Cars using variety of IOT Sensors.
- Technologies Used: Embedded System, Oracle11g, WEB (Front-end, Back-end)
Handwriting Recognition and Documents Parsing & Classification
- The project included the concept of Handwritten Documents sent to HR/ Interview Teams and then Documents such as CV, ID Proofs Parsing, and Classification into Categories.
- Technologies Used: NLP Open-Source Libraries & Toolkit, OpenCV, Flask, Python, SQL, AWS
A system that finds anamoly in the Data.
- This project was based on combination of 3 Algorithms in which the main goal is to keep the system safe from bad data and remove the anomalies from the data and Visualization of the data along with CURD Operations
- Technologies Used:NLP Open-Source Libraries & Toolkit, D3, Flask, Three.js
Awards
Awards & Achivements
Financial Services Hackathon
| Position | 2nd Runner UP |
|---|---|
| Project | Credit Decisioning Using AI |
| Organizer | Accenture |
Project Innovation Contest
| Position | 2nd Runner UP |
|---|---|
| Project | Automation of Automobiles Using IOT |
| Organizer | ICDCIT 2017 |
Best Volunteer
| Award | Best Volunteer 2017 |
|---|---|
| Organizer | National Service Scheme |
| Location | Bhubaneswar |
Patents
Simulation of One or More Pipeline Jobs in a Software Deployment Pipeline
- Patent Issuer and No.: US 18/077, 761
- Patent Filling Date : Dec 29, 2022
Extracting Actionable Items From Documents And Assigning The Actionable Items to Responsible Parties
- Patent Issuer and No.: US 16/525, 172
- Patent Filling Date : Jul 29, 2019
Extracting Actionable Items From Documents And Assigning The Actionable Items to Responsible Parties
- Patent Issuer and No.: AU 2020207818A1
- Patent Filling Date : July 22, 2020
- Patent Grant Date : November 11, 2021
Certification
- Neo4J Professional Certification, Neo4j, 2020
- IBM Professional Data Science, Coursera, 2020
- Cloud Computing Certification, Rackspace, 2016
- Azure AZ-900 Certification, Microsoft, 2020
Education
School of Computer Engineering, KIIT University
Bhubaneswar, India
Degree: Bachelor of Technology in Computer Science