Agentic report generation
Built the agentic AI workflow that auto-generates structured client progress reports from analytics outputs and prepares them for sending. Rolled out company-wide and adopted by the entire analyst team, replacing a manual, per-client report-assembly process.
The bottleneck
Each analyst handles three to four client accounts, and every reporting cycle meant manually assembling a progress report per client from raw analytics outputs — consistent enough in shape to automate, repetitive enough to hurt.
The build
An agentic workflow (Claude, prompt and agent design, Python) that reads analytics outputs and produces a standardised, client-facing progress report ready for review and sending.
Impact
Saves around 30 minutes per client per reporting cycle; across roughly 25–30 accounts company-wide, about 12–15 hours saved every week — with consistent, standardised deliverables across the whole team.
Stack
Interested in work like this?
dhruvnirmal2111@gmail.comOpen to remote data & AI roles — Australia, India, and the Gulf / Singapore.