# AI Upskilling Pathways

Sequences curated AI resources into four enablement stages, one path per team function, ranked by impact so limited time goes furthest.

Status: active

## The Problem

Enterprise AI adoption fails when teams learn generically. A sales team and a product team have fundamentally different AI literacy needs, yet most upskilling programs dump everyone into the same course list. Without structured progression, teams either stall at surface-level awareness or skip straight to tooling without understanding when and why to apply it. The result is scattered adoption — pockets of enthusiasm with no organizational coherence.

## What I Built

AI Upskilling Pathways is a learning management dashboard that organizes curated resources into team-specific paths across four progressive enablement stages:

**Stage 1: Awareness & Literacy** — Foundational understanding of AI capabilities and limitations. Resources from HBR, McKinsey, and Stanford covering what AI can and cannot do, framed for business context rather than technical depth.

**Stage 2: Exploration & Experimentation** — Hands-on exposure to AI tools within safe boundaries. Guided exercises, prompt engineering basics, and evaluation frameworks for assessing AI output quality.

**Stage 3: Integration & Application** — Embedding AI into existing workflows. Process mapping, tool selection criteria, and measurement approaches for tracking productivity impact.

**Stage 4: Optimization & Scaling** — Organizational patterns for scaling AI adoption. Governance frameworks, cross-team knowledge sharing, and continuous improvement cycles.

## Team-Specific Learning Paths

Five paths tailored to different team functions — Research, Strategy, Partnerships, Sales, and Product Management. Each path contains four modules (one per stage) with curated resources. A research analyst's Stage 2 focuses on literature review acceleration and data synthesis tools, while a sales team's Stage 2 covers prospect research automation and proposal drafting.

Each resource carries an impact score (1-5) reflecting the ratio of practical value to time investment. Resources sort by highest impact first so teams with limited time get the most value from whatever they complete.

## Drag-and-Drop Customization

Learning paths are reorderable via drag-and-drop. Team leads can restructure module sequences to match their team's existing knowledge or current priorities. If a product team already has strong AI literacy, they can move Stage 1 modules to the end and start with experimentation.

## Technical Decisions

**Static data architecture** — All learning path content is defined in TypeScript data files rather than a CMS or database. The resource catalog changes infrequently (monthly curation updates), so a database would add complexity without value. Content updates go through code review, which provides version history and quality gates.

**Impact scoring over completion tracking** — Rather than gamifying completion percentages, the dashboard emphasizes impact scores. This shifts the incentive from "finish everything" to "prioritize what matters" — a better model for busy teams with limited learning time.

## Links

- [HTML project page](https://rosslabs.ai/projects/ai-upskill-pathways/)