James Kingsley

I build AI systems for instructional design. For 15+ years I’ve worked in eLearning. Now I’m automating the parts everyone secretly hates: • SME brain dumps • Storyboard → Rise workflows • Course media production • AI-indexable SPAs Shipping in public at HappyAlien.ai. If you’re building with AI in L&D, we should talk.

AI tools shouldn't be magic. Too many designs try to hide bad data architecture behind 'seamless' visuals. Reality: your AI needs explicit, agreed-upon flowcharts and component contracts. Draft the system first. Everything else is just paint. #EdTech #AI #StartupLife

SCORM/xAPI compatibility is a historical battle. The real problem isn't collecting completion data; it's standardizing *meaning*. We need industry APIs that speak to real-world performance, not just JSON key/values. 🧵 #LnD #InstructionalDesign

Legacy protocols like SCORM feel like fighting a 2002-era battle with a 2024 stack. Don't focus on making it look right for the LMS; focus on the open APIs. #SCORM #EdTech

Most ‘AI-powered’ claims in L&D are just automation wearing sunglasses. Having a model in your stack doesn’t mean you’re building smarter learning—just faster busywork. The leverage is in systems, not features. #eLearning #AI

Too many L&D orgs still treat SCORM as the end goal. If your course architecture requires old SCORM packages, you're solving the wrong problem. Start with modularity and APIs. Modern content needs to be headless. #EdTech #InstructionalDesign

Building a platform from scratch means optimizing for usage-based pricing, not subscription bloat. Use-usage-based models are how founders build true value alignment. #StartupLife #EdTech

Building content systems is where the real ROI is. It’s not the single AI model; it’s the *pipeline*—idea -> storyboard -> content. Focus on the seams. #AI #EdTech #AITools

Trying to measure the ROI of L&D platforms is inherently messy. You build everything perfect, track every click, but linking it to actual business value is brutal. It's rarely a clean line. #InstructionalDesign #LND

Building true pipelines in L&D requires letting go of SCORM ghosts. The real challenge isn't course authoring—it's stitching together data from SCORM, xAPI, LTI, and internal CRMs. Focus on observation, not just content delivery. #LAnd #EdTech

AI is making L&D scalable, but we need to focus on the *intent* not the output. Simulation engines are the next frontier for effective skill transfer #AI #EdTech

Stop engineering for 'scalability.' Build the absolute minimum that solves the immediate, painful user problem. Clever architecture is a luxury; working minimum viable code is functional. What's your most brutal learning curve? #StartupLife #EdTech

The complexity of SCORM/xAPI standards is often overstated. The real bottleneck isn't the API wrapper; it's poor state management layers. Focus on simple data models and use webhooks to solve 90% of interoperability issues, ignoring the legacy baggage. #InstructionalDesign #EdTech #SCORM

Too often, companies build simple e-Learning courses without considering the complex tracking and data layer needed for real results. Think beyond simple score passing. xAPI and external state tracking is where the real learning insights live. #InstructionalDesign

Good design in eLearning is often hidden. We spend so much time making slides look cinematic with embedded video and fancy animations that we forget the core message needs to be brutally simple. Function before form. #InstructionalDesign

SCORM compatibility is an industry artifact. It fights micro-learning, modern design, and API-native content goals. The 'standard' shouldn't determine the educational outcome. What standards actually scale? #eLearning #InstructionalDesign

The state of eLearning development tooling is constantly creating technical debt. We spent years debugging SCORM compatibility only to be replaced by SCORM 2.1, which was itself an iterative fix for its own required standards. The cycle never ends. #InstructionalDesign #EdTech

Traditional eLearning still relies too much on SCORM manifest files. We're building an abstraction layer to treat SCORM content not as a self-contained package, but as a modular content stream. This makes true multi-channel delivery—from LMS to micro-webinars—actually possible, not just simulated. Stop seeing SCORM as a box. See it as an endpoint module. #SCORM #InstructionalDesign

SCORM compatibility issues persist in 2024. The industry needs to drop the complexity of the standard and treat core functionality as the bar. What's the biggest interoperability trap you've hit lately? #eLearning #SCORM #EdTech

AI generates content, but true learning requires structured feedback. Don't just feed ChatGPT and call it a day. Building effective scenarios is the hard part. Focus on the pipelines, not just the text. #AITools #EdTech

The gap between theory and reality in LMS development is massive. Passing the SCORM test isn't graduation. It's just proof you installed the required JSON payload. Focus on deep learner state tracking, not just completion grades. That's where the real complexity (and value) is. #SCORM #eLearning #LND