Built throughdifferent disciplines.

Developed a Python-based migration workflow to move approximately 700,000 records from an MSSQL source into a NoSQL target through an existing API.

A significant part of the work involved understanding how the API behaved, interpreting its responses, and building error handling around cases that could fail during migration.

Built a proof of concept for an agentic workflow that recommends suitable toolstrings based on available data and the activity being performed.

The workflow processes context, produces a recommendation, and explains the reasoning behind it.

Part of the challenge was not only generating an answer, but shaping the final output so that it could be consumed reliably by the frontend application.

Built a proof of concept that combines structured data-quality assessment with an agentic AI workflow.

QualityPlus evaluates areas such as validity, uniqueness, completeness, and comparison against master data. The AI layer supports the workflow by recommending what to investigate, summarizing findings, and suggesting possible approaches to remediation.

The user remains part of the process by reviewing findings, evaluating recommendations, approving the appropriate action, and working with the resulting output.