V Krishna Kumar

Senior Analyst · Data & Automation

Chennai, India | krrish1606@gmail.com | (+91) 9094952806 | linkedin.com/in/mrvkrishnakumar

Professional Summary

Senior Analyst with about 11 years of experience in data reporting, automation, and AWS cloud engineering. I’ve completed the AWS Certified Solutions Architect – Associate and AWS Certified Cloud Practitioner exams. At Empower, I build serverless pipelines that process retirement plan documents, using Python, AWS, SQL, and LLMs, saving hundreds of manual hours a day. Earlier, at TCS, I worked on SAP BusinessObjects and the Information Design Tool.

Skills & Expertise

Cloud & Serverless (AWS): AWS Lambda, Step Functions, API Gateway, DynamoDB, Amazon S3, Amazon Textract, EventBridge, SQS, CloudWatch, CloudFormation, Secrets Manager, Redshift

AI & Automation: Python, LLM Integration (APIs), LangChain, Human-in-the-Loop (HITL), Microsoft Graph API, Workflow Automation

Data & Analytics: SQL, SAS, SAP Web Intelligence, Tableau

Work Experience

Senior Analyst

Aug 2023 – Present
Empower | Bangalore, IN
  • Built a serverless pipeline on AWS using Lambda, Step Functions, Textract, and DynamoDB. It auto-indexes retirement plan documents by filling in the matching Hyland OnBase index fields, such as SSN, account number, and customer name, straight from each PDF or TIFF. Low-confidence cases go to a human reviewer, which removed hundreds of manual indexing hours a day.
  • Added LLaMA 3.3 (70B) to the pipeline to find the customer's account number. Textract sometimes returned both the customer's and the beneficiary's account numbers, and the model picks the right one. Its output is then checked against the database.
  • Automated payment registry processing. SAS combines data from multiple sources into several Excel files, then merges and formats them and sends an automated email when the report is delivered. This saves 30 minutes of manual work a day.
  • Automated email extraction through the Graph API. The body of each email is pulled out, turned into an Excel sheet, and saved to a network location. This saves 2 hours a day in reporting.
  • Used Amazon Textract to convert table data in PDFs into Excel, saving 1 hour a day. Added a human review step so people approve the data before the database is updated.
  • Designed an AWS multi-region active-hot standby architecture for high availability and disaster recovery, with minimal failover delay. Built a CloudWatch dashboard with two buttons, us-east-1 and us-west-2. If us-east-1 goes down, the process can be run from us-west-2 by clicking the button, and vice versa.
  • Built SQL validations for loan document processing. Textract and Python pull the loan amount, account number, and participant name from each document, then compare them against the Oracle database. Applications either move forward for further processing or go to manual review. This cut verification time from hours to minutes.
  • Built a CloudWatch metrics system and dashboard to monitor Lambda, S3, and Step Functions health, with Log Insights.
  • Designed a serverless ETL pipeline on AWS for Pega case data. Step Functions and EventBridge orchestrated several Lambda jobs that extracted, transformed, and loaded the data into a Redshift table through S3 staging, then published it as a Tableau Hyper extract. Built the architecture with concurrency guards, SQS dead-letter queue, and CloudWatch alarms for failure alerts. The pipeline ran every 30 minutes, pulling the latest data and refreshing the Hyper extract.

Analyst

Jul 2021 – Aug 2023
Empower | Bangalore, IN
  • Automated tax reporting using SAS and Python, saving 45 min/day and $30,000 annually in vendor costs.
  • Automated email deletion, Salesforce request creation, and attachment downloads — saving cumulative hours weekly.
  • Created multiple SAS reports improving data accessibility and decision-making across teams.
  • Built a Python script that uses Selenium to log into Salesforce, download reports as Excel files, and place them in a shared folder, this saved 1 hour of manual effort daily in downloading the reports.

Intern

May 2020 – Sep 2020
Empower | Bangalore, IN
  • Built topic modeling + logistic regression pipeline (80% accuracy) to classify unstructured NIGO form text.
  • Developed Tableau dashboard to visualize rejection reasons and provide actionable recommendations.

IT Analyst

Nov 2014 – Aug 2019
TCS | Chennai, IN
  • Owned the reporting work end to end, from gathering requirements and building the semantic layer to designing reports in SAP Business Objects.
  • Integrated Oracle and MS-SQL data into new reports built to the requirements.
  • Supervised 2 associates, reviewing their work and keeping their deliverables on track.
  • Automated Timesheet report generation, saving 15 minutes a day.
  • Built an incident data dashboard for management presentations.
  • Scheduled reports in Infoburst and made sure they ran on time and reached users.

Education

B.Tech – Mechanical Engineering

2010 – 2014  ·  CGPA: 9.066 / 10
SRM University | Chennai, IN

Certifications

Awards

Personal Projects