Overview
Jesse. was born from a simple but costly truth: paperwork slows down freight. Drivers waste hours scanning, resending, or chasing down documents, while brokers and warehouses absorb the inefficiencies in the form of disputes, delays, and lost revenue.
As the founding designer and product lead, I set the vision for Jesse., an AI-powered paperwork assistant that transforms fragmented processes into a seamless workflow. This case study captures how I defined the problem, led research with real drivers and brokers, translated findings into a product strategy, and delivered a production-ready MVP using Cursor and Supabase.
My Roles
Founder & Product Designer — Established the product vision and roadmap, anchored in both user insights and market gaps.
UX Strategy & Research Lead — Directed qualitative research in driver communities and synthesized findings into actionable product principles.
Design System Architect — Created the brand, interface patterns, and a scalable design language aligned with driver-first simplicity.
Low-Code Builder — Delivered a working MVP with Cursor + Supabase, handling data modeling, AI integration, and end-to-end flow design.
Tools & Stack
Research & Strategy: Notion, Discord, Reddit forums, direct interviews
Design: Figma, Artboard Studio (design systems, interaction flows)
Build: Cursor, Supabase (auth, storage, database)
AI Layer: Claude Code + GPT for reasoning and document cleanup
Automation: Zapier, Airtable for operational workflows
The Problem
Freight logistics runs on documents — bills of lading, proofs of delivery, receipts — but the workflows around them are outdated. Through research and stakeholder interviews, I mapped out three recurring breakdowns:
Drivers lose 2–3 hours/day redoing paperwork, directly reducing their earning potential.
Brokers spend hours fixing errors; nearly 25% of invoices require corrections.
Warehouses face 15% dock inefficiency due to missing or mismatched documents.
The “solutions” on the market (scanner apps, PDFs, email) were built for office workers — not drivers, pressed for time and operating with patchy connectivity.
Research & Discovery
I conducted research across Discord groups, Reddit and trucking forums, and direct driver interviews to understand workflows end-to-end. My goal wasn’t just to identify pain points, but to uncover the deeper motivations that would guide product strategy.
Key insights:
Drivers don’t care about “document management” — they care about getting paid faster.
Brokers need consistency: clean, formatted packets they can invoice without dispute.
Simplicity is greater than features.
These insights became product principles: optimize for clarity, automate formatting, and reduce touch points to the absolute minimum.
The Solution
The Jesse. MVP operationalizes the vision through a streamlined workflow:
Snap a Photo – Jesse. auto-enhances, crops, and prepares the doc.
Jesse. Cleanup + Formatting – OCR and AI normalize data into broker-ready packets.
Send to Broker – Instantly share standardized documents, eliminating disputes and delays.
The design emphasizes speed, trust, and transparency. Drivers remain in control — they can review before sending — but Jesse. handles the complexity in the background.
UX Process
I led the design process from concept to delivery:
Mapping the workflow — Documented the driver journey from load pick-up to broker payment, identifying decision points and failure modes.
Lo-fi explorations — Explored conversational onboarding, one-tap workflows, and preview-verification flows.
Prototyping and feedback loops — Shared prototypes back into driver forums, refining based on real feedback.
Critical design decisions:
Pivoted from an early voice-driven concept (abandoned due to low trust) to a photo-scan workflow with offline-first sync.
Simplified navigation into three core tabs (Camera, Docs, Contacts) for clarity and speed.
Built an MVP-ready design system with scalable UI tokens and repeatable interaction patterns.
Reflection
