AI Systems & Case Studies

Problem: Attorneys often start drafting contracts from client intake data, then spend significant time re-entering facts, organizing clauses, and producing a consistent first draft.

Solution: A structured AI workflow normalizes intake data, assembles controlled prompts and templates, generates a first draft, and routes it to an attorney for review and approval.

Role: AI workflow design, prompt engineering, system integration, testing, and product thinking

Outcome: Created a repeatable drafting process that reduces repetitive first-draft work while preserving attorney control over legal judgment and final content.

Problem: Group sales lead handling required manual discovery, routing, availability checks, scoring, and draft outreach before a sales manager could act.

Solution: A multi-agent workflow writes structured outputs into a shared data layer and presents them in a dashboard designed for sales review and action.

Role: Product design, dashboard implementation, workflow integration, Airia trigger design, Supabase integration, and stakeholder walkthrough design

Outcome: Reduced a manual workflow from hours to under 10 minutes while keeping final review in human hands.

Problem: Families seeking disability services often face dense policy language, limited accessibility, and slow content updates when regulations change.

Solution: A RAG-based pipeline grounds scripts in verified policy documents, routes content through human review, translates approved scripts, and prepares multilingual video explainers.

Role: AI systems product designer responsible for workflow architecture, RAG pipeline design, governance flow, and human review checkpoints

Outcome: Demonstrated how AI can responsibly scale public-sector education while keeping policy accuracy and human approval central to the workflow.

Problem: Photography session operations required repeated manual follow-up across session status, location details, gallery delivery, favorites selection, editing timelines, and final delivery communication.

Solution: A Supabase-backed admin dashboard and client portal organize sessions into a status-based workflow with smart delivery estimates, client actions, and automated email triggers.

Role: Product designer, full-stack builder, workflow architect, and AI-assisted development lead

Outcome: Improved operational clarity, client communication, delivery timeline tracking, and studio professionalism while preserving personal client touchpoints.

Problem: Music chart data is dense, fragmented, and difficult to interpret quickly when users have to cross-reference static reports or BI dashboards.

Solution: A custom web analytics experience separates chart position, performance drivers, comparative rankings, and conversational data exploration.

Role: Product engineer responsible for dashboard design, full-stack implementation, data modeling, and AI-assisted exploration patterns

Outcome: Reduced chart interpretation from minutes of reading to seconds of scanning and enabled ad-hoc questions that previously required custom views.

Problem: Online sports leagues require repeated manual work for scheduling, matchup setup, rosters, records, and community administration.

Solution: A Discord-native SaaS platform automates league operations, connects subscriptions and license delivery, and uses AI-assisted image extraction to reduce manual schedule entry.

Role: Founder, product lead, and product engineer responsible for product strategy, full-stack implementation, AI workflow design, and go-to-market flow

Outcome: Reduced commissioner workload through automated league workflows, AI-assisted data ingestion, and a subscription-backed product experience.