Dynamic RPA Engineer at New Vision Softcom & Consultancy, specializing in process automation and data analysis. Expertly developed and deployed a bot that optimized invoice processing, significantly enhancing efficiency and accuracy. Demonstrated capability to collaborate with teams and resolve technical challenges, driving substantial improvements in operational workflows.
Overview
1
1
year of professional experience
1
1
Certification
Work History
RPA Engineer
new vision softcom & consultancy
Hyderabad , Telangana
06.2021 - 10.2022
Process Discovery and Analysis: The Detective Phase. This is the most critical step. Automating a bad process just gives you a faster, bad process.
Process Mining: Used specialized tools to analyze user activity logs to objectively identify the most repetitive and time-consuming tasks.
Bot Development and Scripting (The Builder Phase): This is where the actual automation is built.
Tool Selection: I worked primarily with industry-leading RPA platforms like UiPath and Blue Prism.
I designed the automation workflow visually, mapping out every click, keystroke, data entry, and decision point.
A Concrete Example: Automating Invoice Processing. Let's say the accounts payable team was manually processing hundreds of PDF invoices per day. My project would look like this:
1. Discover: I confirm with the team that this is a repetitive, rule-based task taking 5 or more FTEs (Full-Time Employees).
2. Analyze: The process is: receive email -> download PDF -> extract data (vendor name, invoice #, amount) -> enter data into SAP -> file the PDF.
3. Build: I develop a bot using UiPath that: * Logs into the dedicated email inbox.
Downloads PDF attachments.
Uses OCR (Optical Character Recognition) to read the invoices.
Extracts the key data fields into a structured format.
Logs into SAP and inputs the data into the correct fields.
Saves the PDF to a network drive with a specific naming convention.
Sends a "for approval" email if the amount is over a certain threshold (an exception).
4. Test: I run the bot on a sample of 100 old invoices and compare its output to the known correct data to ensure 99.5%+ accuracy.
5. Deploy: I schedule the bot to run every hour on a dedicated virtual machine. The AP team now only needs to handle the exceptions that the bot flags.