Sairam Siripuram

Improve insight readiness for Sindalah stakeholders by ingesting multi-source guest data.

Enable accurate predictions by processing ingested data through custom Spark jobs.

Reduce downstream errors by applying business rules that isolate invalid records.

• Strengthen data understanding by modelling entity relationships across Postgres and Iceberg.
• Increase query flexibility by storing metadata for a semantic data virtualisation layer.
• Accelerate defect resolution by partnering with assurance team during quality testing.

Developed a data-operations hand-off shift portal using Python with Django capabilities.

Owned model, view, and template design to streamline user request handling.

Drove data cleansing and ingestion into NDM with consistent structuring.

• Chose Python, AWS, and Jenkins to automate distribution workflows and handovers.
• Set up web applications to track custom builds and report build status.
• Directed Hadoop report automation to monitor directories above 90% storage.
• Implemented automated retention controls to remove data older than 3 months.

Implemented proactive monitoring for suspicious purchasing patterns across organisation-wide reporting lines.

• Collaborated with legal and procurement to design monitoring rules and reporting.
• Deployed spendthrift rules for fraud detection and expense reimbursement workflows.

Manage 302 certification application quarterly to support SHCS SOX compliance control.

Built internal portal with leave, investments, user, payroll, and procurement modules.

Developed MIV consumption and OP reports with upload and update features.

Designed and maintained audit issue tracking with Solar search integration.

Gathered client requirements and prioritised portal enhancements for CSOD and Dunkin.

Owned development of new portal functionalities for CSOD and Dunkin Brands.

Coordinated validation testing to verify features matched agreed requirements.

Analysed 2 billion CDR records, uncovering close-knit suspicious communities for further investigation.

Constructed community detection model using telecom CDR topology and contact features to identify patterns.

Developed ad server model predicting user home-location to enhance targeted advertising strategies.