Toledo Athletics Onboarding Portal
A serverless internal platform that centralizes onboarding knowledge, policies, contacts, org chart data, systems directories, and moderated staff tips for University of Toledo Athletics staff.
Onboarding is a knowledge-management problem
The Toledo Athletics Onboarding Portal is an internal web platform built for the University of Toledo Athletics Department. Its purpose is to help new employees quickly understand the department's structure, policies, tools, compliance expectations, and day-to-day operating environment — without relying on scattered documents, email chains, and informal conversations.
A Division I athletics department is not a normal workplace. New staff must navigate university HR, NCAA compliance, Title IX, NIL rules, student-athlete support systems, brand standards, facilities, ticketing, development, communications, IT systems, campus logistics, and athletics-specific workflows simultaneously.
The portal turns that fragmented institutional knowledge into a centralized, searchable, structured internal resource. It combines official published articles with moderated employee-submitted tips so the department can capture practical knowledge without allowing unverified information to become official guidance.
New staff face a steep learning curve in a high-compliance environment
Incoming athletics staff need answers to questions that span a dozen different functional areas — often before anyone has had time to explain them. Without a centralized system, onboarding becomes inconsistent: new employees waste time searching for answers, supervisors repeat the same explanations, and outdated information can create operational or compliance risk.
NCAA, Title IX, NIL, FERPA, and HR rules require timely understanding. Incorrect or missing guidance creates institutional risk.
Critical information lives across web pages, PDFs, emails, internal documents, and informal conversations with no single entry point.
Without a reference system, supervisors repeat the same explanations for every new hire — an operational tax that compounds across a large department.
Different hiring managers provide different information, producing uneven readiness across staff in the same role.
Static documents go stale as policies update, staff move roles, systems change, and leadership changes — but there is no mechanism to surface or fix them.
Practical wisdom from experienced staff — the kind learned only by doing the job — has no home. When staff leave, it leaves with them.
The onboarding portal addresses all of these by creating a single maintained source of truth — with a moderated workflow that allows staff knowledge to flow in without bypassing compliance-safe content control.
A moderated knowledge platform, not an open wiki
The portal is built around three distinct layers of content. The first is official centralized knowledge: long-form onboarding articles, policy resources, HR timelines, compliance references, university systems, and athletics-specific guidance. The second is structured operational reference data: org chart, key contacts, quick links, systems directory, branding tokens, and policy resources. The third is moderated employee knowledge captured through a controlled tips workflow.
Moderated tips workflow
This design decision matters most. An open internal wiki where anyone can publish unverified guidance is dangerous in a compliance-heavy environment. The portal instead implements a controlled knowledge pipeline:
Staff member submits a tip, advice item, or suggested content update through the portal.
Submission is stored as pending and is not visible in the live portal.
A moderator or administrator reviews the content for accuracy and relevance.
The reviewer approves, rejects, or requests revisions. Only approved content becomes visible.
Published content can receive feedback if it becomes outdated — routing it back into the review queue.
Content areas
Serverless full-stack — lightweight but real
A purely static site would not have been sufficient. The platform needed form submissions, moderation states, user roles, relational data, content updates, AI chat, database-backed directories, and future approval workflows. Cloudflare Workers and D1 provide the backend and database layer required for a real internal platform while staying lightweight and cost-effective.
System flow
14 tables separating official content from unverified submissions
The schema is designed to enforce a clear boundary between official content and staff-submitted knowledge. Content can evolve without rewriting the application — staff, policies, links, systems, and articles can all be updated through seed files.
The SiteContentIndex table powers AI and search retrieval across published content. The Submissions and Tips tables enforce the moderation boundary — unapproved content never surfaces in the live portal.
Navigation aid, not a policy authority
The portal includes an AI chat assistant powered by Cloudflare Workers AI. It is designed to improve discoverability — helping users find information faster without replacing the need to verify compliance-sensitive decisions against official sources.
Designed to answer questions like
Using a native Workers AI binding requires no external API key — the AI runs in the same Cloudflare edge environment as the rest of the stack, keeping the platform simple to deploy and operate. The SiteContentIndex table provides the retrieval layer that grounds the assistant in actual portal content rather than general model knowledge.
Technical stack
Internal tools require product thinking, not just code
The challenge was not writing a website. It was deciding what information belongs in the portal, how it should be structured, what can be staff-contributed versus what must be official, and how to keep it maintained when the department changes.
Athletics onboarding covers NCAA rules, Title IX, FERPA, NIL, and HR policy — areas where incorrect guidance creates institutional risk. A moderation layer is not bureaucracy; it is a safety control.
Cloudflare Workers and D1 provide the dynamic capabilities a real internal tool needs — form submissions, database queries, moderation states, AI chat — without the overhead of a managed server or a heavy framework.
Hardcoding onboarding content into pages would make the site fragile. A relational schema means content — staff, policies, links, systems, articles — can be updated through seed files without changing application logic.
Onboarding content changes frequently. A portal that cannot be maintained becomes outdated and untrustworthy. The maintenance guide designed alongside the platform is as important as the code.
An AI assistant that answers from general model knowledge is unreliable for policy and compliance questions. Connecting the assistant to the SiteContentIndex — a structured index of published portal content — makes answers more trustworthy and correctable.
A foundation for a full department operating system
- Authenticated UToledo SSO login
- Role-based admin dashboard for content approval
- Onboarding progress tracking for each new employee
- Manager checklists for 30/60/90-day reviews
- Semantic search across articles and policies
- Analytics for most-viewed and unanswered questions
- Automatic stale-content reminders
- CSV import/export for staff and contacts
- AI-generated summaries for long policy documents
- Department-specific onboarding paths (coaches, operations, communications, compliance, ticketing, development)
- Audit logs for all content updates
- Mobile-first quick-start mode for first-day onboarding
- Cross-role knowledge transfer workflows when staff change positions
Great onboarding is infrastructure for institutional memory.
The Toledo Athletics Onboarding Portal demonstrates that internal software engineering — serverless architecture, relational schema design, moderated knowledge workflows, and AI integration — can directly reduce operational risk and help people do their jobs better from day one.