Grant Automation

Production-grade, audit-ready pipelines for nonprofit grant management and compliance reporting.
Deterministic parsing over probabilistic guessing · bounded contexts over glue code · immutable audit trails over best-effort logs.

www.grantautomation.org

---

What this is

Grant Automation is a deep, engineering-first reference for the people who keep nonprofits out of audit findings: grant managers, nonprofit operations leads, compliance officers, and the Python developers who build the automation behind them.

Every page is opinionated and copy-ready. The Python examples use type hints, Pydantic validation, structured logging, and explicit audit hooks; the compliance guidance cites the actual regulation — 2 CFR §200.302, §200.414, §200.334, IRS Form 990 parts and schedules, FASB ASC 958, and NIST SP 800-53 controls — rather than hand-waving at "compliance requirements." The patterns are written to survive an external audit and a midnight pipeline failure with equal grace.

What you'll find

The site is organised into two in-depth references, each broken into focused, hands-on guides:

• Data Ingestion & Grant Parsing Workflows — deterministic multi-format extraction for PDFs, spreadsheets, APIs, and webhooks. Canonical intermediate representations, strict schema validation, normalization, batch pipelines, and structured error handling.
• Core Architecture & Compliance Mapping — bounded contexts, deterministic rule evaluation, jurisdictional separation (federal / state / grantor), audit-trail evidence automation, indirect-cost-rate handling, and audit-ready artifact generation.

A few representative walkthroughs:

• Extracting tables from grant PDFs with pypdf and Camelot
• Camelot vs pdfplumber for grant budget tables
• Verifying HMAC signatures on grant portal webhooks
• Building an append-only audit ledger in Postgres
• Automating 2 CFR 200 record retention schedules
• Mapping IRS 990 Part VII to a JSON schema
• End-to-end grant pipeline walkthrough

Why it's different

• Deterministic by design. Same input, same output, every run — the only foundation an auditor can replay.
• Regulation-specific. Guidance is mapped to named citations and audit artifacts, not generic advice.
• Audit-ready. Immutable, hash-chained, cryptographically signable evidence with retention baked in.
• Genuinely hands-on. Runnable Python, schema contracts, pytest/hypothesis tests, and hand-authored diagrams on every page.

Tech

This is a static site built with Eleventy. Content is Markdown with structured front matter; pages are enriched with JSON-LD (TechArticle, BreadcrumbList, HowTo, FAQPage), custom accessible inline SVG diagrams, and a small, dependency-free client. It deploys to Cloudflare.

Build locally

npm install
npm run build      # generates the static site into _site/
npm run serve      # local dev server with live reload

Deploy

npm run deploy     # build + publish to Cloudflare

Contributing & contact

Issues and suggestions are welcome via the GitHub repository. If a regulatory citation drifts or a code sample needs updating for a new library release, open an issue.

_{© Grant Automation · www.grantautomation.org}