PAMELAi™

What Is PAMELAi™?

PAMELAi™ is an enterprise SaaS platform purpose-built for the US and Canadian investor-owned utility (IOU) sector. It combines regulatory docket intelligence, rate case precedent analysis, return-on-equity (ROE) benchmarking, and AI-powered regulatory data collection into a single, unified interface designed for utility executives, rate case teams, and regulatory law firms.

The platform targets the approximately 197 investor-owned utilities across the US and Canada — the electric, natural gas, and water utilities that operate under state and provincial regulatory oversight and must file for rate increases, capital recovery, and earnings authorizations through formal regulatory proceedings (rate cases) before public utility commissions.

The Three Pillars

The Problem PAMELAi™ Solves

Regulatory proceedings for US investor-owned utilities involve thousands of dockets across 50+ commissions, spanning decades of rate case decisions. Lawyers, analysts, and utility executives face a fragmented, time-consuming research landscape:

Platform Architecture

PAMELAi™ is a modern full-stack TypeScript application with a React frontend, Express backend, and PostgreSQL database. The entire platform runs as a single deployable unit on Replit cloud infrastructure.

Key Files & Their Roles

PAMELAi/
├── client/src/
│   ├── pages/
│   │   ├── admin.tsx          ← Full admin panel (RBAC, scrapers, testing, quality)
│   │   ├── law-firm.tsx       ← Docket Tracker, ROE Workbench, Precedent DB
│   │   └── home.tsx           ← Public landing page
│   └── App.tsx                ← Routing + auth guard
├── server/
│   ├── index.ts               ← Express app entry, middleware, scheduler wiring
│   ├── routes.ts              ← Main API routes (dockets, utilities, ROE history)
│   ├── law-firm-routes.ts     ← Law firm module API (precedents, benchmarks, PDF)
│   ├── scrapers.ts            ← AI scraper engine (58-commission registry + GPT-4o)
│   ├── test-suite.ts          ← 42-test automated test suite (8 suites)
│   ├── precedent-pipeline.ts  ← Seed + normalize precedent records from ROE history
│   └── storage.ts             ← Database interface (IStorage) — all CRUD operations
├── shared/
│   └── schema.ts              ← Drizzle ORM schema (all tables + types)
└── client/public/
    └── PAMELAi-Platform-Guide.html  ← This document

Database Schema

Authentication & Authorization

Feature-by-Feature Breakdown

🔍 Docket Tracker

The Docket Tracker is the operational hub for monitoring active rate cases. It ingests docket filings from the built-in scrapers and lets users filter, search, and track proceedings by utility, state, docket type, and status.

• Filter by state, utility type (Electric/Gas/Water), status (Active/Filed/Closed), and jurisdiction
• Search by docket number or utility name — instant results across the full database
• Add dockets manually — with utility name, docket number, jurisdiction, commission, type, and status
• Auto-populated from scrapers — built-in scrapers add new dockets daily from FERC, CPUC, FPSC, ICC, NYPSC, PUCT
• Export to PDF — benchmark report generation for regulatory filings

📚 Precedent Database

The Precedent Database stores historical rate case decisions with their authorized ROE, capital structure, and jurisdiction. It is the research backbone for rate case attorneys preparing testimony on peer utility comparisons.

📈 ROE Expert Witness Workbench

The ROE Expert Witness Workbench is PAMELAi™'s flagship analysis tool — a full in-platform financial modeling suite designed for rate case expert witnesses and regulatory analysts. It combines peer group benchmarking with three independent cost-of-equity estimation models and generates ready-to-file exhibit language.

🤖 AI Scraper Engine

The most technically distinctive feature of PAMELAi™. Rather than requiring manual configuration for each of the 58 regulatory commission websites (all with different URL structures, search forms, and formats), the AI Scraper Engine uses GPT-4o to:

1. Look up each commission in the registry by name, state, and known URL
2. Infer the correct API endpoint or search URL for docket listings
3. Generate a scraper configuration including selectors, pagination strategy, and data field mappings
4. Save the config to the database for reuse on every daily scrape cycle

What PAMELAi™ Can and Cannot Do

The following is an honest, complete inventory of platform capabilities as of March 2026 (Beta). Features marked with △ Planned are on the roadmap.

Industry Comparison

PAMELAi™ occupies a specific niche: utility regulatory intelligence for rate case teams. It is not a general legal research tool, not a financial data terminal, and not a document management system. The following compares it to the tools that teams currently cobble together to solve the same problems.

Narrative Summary

Best-in-Class Score by Category

Complete Tutorial — Using PAMELAi™

Part A — Getting Started

Part B — Tracking Dockets

Part C — Researching ROE Precedents

Admin Panel — Complete Guide

Admin Panel Tabs

Scraper Automation Tab — Step by Step

Testing & Data Quality

PAMELAi™ is one of the only vertical SaaS platforms in the utility regulatory space to ship with a built-in automated test suite and data integrity scanner. This reflects the high-stakes nature of regulatory data — a wrong ROE value or corrupted state code in a regulatory filing can have legal consequences.

The 42-Test Automated Suite

The test suite runs in ~112ms and covers 8 suites:

Running the Test Suite

Data Quality Scanner — 10 Automated Checks

Best-in-Class Verdict

This is an honest assessment. PAMELAi™ is early-stage. Here is where it leads the market and where it needs to grow.

When to Choose PAMELAi™

• You are a law firm or utility regulatory team needing a centralized docket + ROE tool
• You want self-serve access without a months-long enterprise sales process
• You need AI-powered scraper coverage across many commissions without a dedicated data engineering team
• You are tired of maintaining manual Excel-based ROE comparables spreadsheets
• Cost matters — you cannot justify $100K+/yr for S&P Global

When to Choose a Competitor Instead

• You need decades of historical rate case archives (S&P/RRA is better)
• You need SOC 2 certified data handling (use an established vendor)
• You need general legal research beyond utility regulatory (Bloomberg Law or Westlaw)
• You need enterprise SSO + multi-tenant data isolation today (roadmap, not yet live)

Pricing & Access

Product Roadmap

API Reference

All endpoints use JSON. Admin endpoints require Authorization: Bearer <SESSION_SECRET>. User endpoints require a valid Replit OIDC session cookie.

GET  /api/law-firm/dockets          → List dockets (filter: state, type, status, search)
POST /api/law-firm/dockets          → Create docket
GET  /api/law-firm/precedents       → List precedents (filter: state, type, year, search)
POST /api/law-firm/precedents       → Create precedent
GET  /api/law-firm/roe-history      → ROE history (filter: state, year, type)
POST /api/law-firm/roe-history      → Add ROE history entry
POST /api/pdf/benchmark             → Generate PDF benchmark report

GET  /api/admin/scrapers/configs          → List dynamic scraper configs
POST /api/admin/scrapers/configs          → Create new config
PATCH /api/admin/scrapers/configs/:id     → Update config (enable/disable/edit)
DELETE /api/admin/scrapers/configs/:id   → Delete config
GET  /api/admin/scrapers/registry         → All 58 commissions with coverage status
POST /api/admin/scrapers/discover         → AI auto-build configs for missing commissions
POST /api/admin/scrapers/run/:source      → Manually run a specific scraper
GET  /api/admin/scrapers/runs             → Scraper run history

GET  /api/law-firm/peer-groups            → List all saved peer groups
POST /api/law-firm/peer-groups            → Save a named peer group (members + workbench_params)
DELETE /api/law-firm/peer-groups/:id      → Delete a peer group
GET  /api/utilities                       → Full IOU utility registry (filter: sector=electric|gas|water)

POST /api/admin/testing/run               → Run all 42 tests (8 suites), save results
GET  /api/admin/testing/runs              → Test run history (latest first)
GET  /api/admin/testing/runs/:id          → Single run with full per-test results
GET  /api/admin/testing/tests             → All 42 tests listed (no execution)
POST /api/admin/testing/scan-data-quality → Run 10-check data integrity scan

GET  /api/auth/user     → Current user (or 401 if not signed in)
GET  /api/login         → Redirect to Replit OIDC login
GET  /api/logout        → Clear session, redirect to home