Case Studies
Selected work showing how I turn manual processes, scattered information, and fragile implementation details into clearer systems people can use, inspect, and improve.
Problem: practical automation content can be hard to trust when readers cannot inspect the actual workflow.
System built: a public n8n workflow library with tutorials, scrubbed GitHub repos, documentation, screenshots, and support articles.
Role: designed, built, documented, packaged, and connected the workflows into GetPrompting's automation education system.
View Case StudyProblem: notes, tasks, inbox checks, content tools, local models, and media workflows were spread across too many places.
System built: one visible operating layer for planning, retrieval, content prep, guarded coding, documents, review, and local AI services.
Outcome: an estimated 3 to 6 hours per week reclaimed by reducing repeated setup, context retrieval, and coordination work.
Explore the Pixel SystemProblem: working notes were useful but too messy to become a reliable daily planning artifact.
System built: a reusable n8n workflow that turns unstructured notes into a clean Google Doc action brief.
Learning: small workflows are easier to review, package, teach, and adapt than large systems with unclear inputs.
View Case StudyProblem: a complex WordPress presentation layer needed cleanup before it felt steady enough for real visitors and stakeholder review.
System improved: page structure, design consistency, interactive graph QA, readability, custom wrappers, and review flow.
Learning: strong implementation often means respecting the existing material while making the system around it easier to trust.
View Case StudyProblem: keyword research was taking too much manual cleanup before it could support useful content decisions.
System built: an n8n workflow that collects, cleans, deduplicates, and structures keyword data using OpenAI, DataForSEO, and Google Sheets.
Learning: automation is most useful when it produces cleaner planning data, not just more output.
View Case StudyProblem: AI workflow experiments need grounded knowledge and enough structure to debug where an answer came from.
System built: a modular retrieval foundation for ingestion, document normalization, metadata, embeddings, vector retrieval, and local model experimentation.
Learning: AI systems become more useful when they are maintainable workflow infrastructure, not just a chatbot interface.
View Case StudyNext Step
If your team has a recurring process that is too manual, scattered, or hard to reuse, we can start with one focused workflow and decide whether an audit, prototype, or documentation cleanup is the right next move.
Start a Consulting Conversation