LLM Project Portfolio - David Klan
EN | 日本語

Project Detail

Enterprise AI Enablement in Insurance – Reporting & Incident Intelligence

Automated reporting and incident intelligence to improve executive visibility and decision making.

  • Enterprise Automation
  • Incident Intelligence
  • Executive Reporting

Executive Summary

Problem

PM reporting and incident trend analysis were fragmented across enterprise tools, limiting executive visibility and slowing Change Management response.

Solution

Automated reporting pipelines and built executive dashboards, then introduced bespoke parser workflows (with coding agents) to analyze large incident datasets.

Outcome

Improved reporting cadence, clearer incident trend signals, and stronger evidence for Change Management decisions.

Use Case & Stakeholders

Insurance leadership needed reliable, cross-system intelligence for delivery status, incident patterns, and operational risk.

  • Executive leadership
  • Program and project managers
  • Change Management and incident response teams

Architecture

Operational data from collaboration and ticketing systems is transformed into dashboard-ready metrics, with parser pipelines for deep incident trend extraction.

[Confluence | Jira | SharePoint | ServiceNow]
             |
   Automation + ETL orchestration
             |
   Power BI executive dashboards
             |
 Incident data parser pipelines
             |
 Change Management trend insights

Tech Details

RAG / Prompt Patterns

  • Structured summarization prompts for status rollups.
  • Parser-assisted analysis prompts for incident clustering.
  • Executive-tailored narrative templates for dashboard briefings.

Tools

  • Confluence, Jira, SharePoint
  • Power Automate and Power BI
  • Bespoke parser scripts supported by coding agents

Constraints

  • Data quality inconsistency across tools and teams.
  • Large incident corpora requiring efficient parsing.
  • Need for executive-level clarity without losing technical detail.

Tradeoffs

  • Highly automated reporting improves speed but needs governance checks.
  • Deep parser analysis increases insight quality but requires stronger maintenance.

Screenshots Gallery

Lessons Learned / What I'd Improve Next

Lessons Learned

  • Garbage in = Garbage out. Data quality is critical for automated reporting success.
  • Data format consistency is essential for reliable parsing and analysis.
  • Executive communication improves when metrics and narrative are co-designed.

What I'd Improve Next

  • Add Power Automate workflows to handle document processing.
  • Add quality gates for parser output validation.
  • Improve dashboard focus for actionable insights.

Repro Notes / Demo Walkthrough

Non-runnable public demo: use the walkthrough steps below.

  1. Setup processes to capture data and receive feedback from surveys.
  2. Link to Power BI dashboards and validate KPI accuracy.
  3. Use ML to detect data trends and anomalies for incident and problem reduction.