# Gil Medrano > I help startups and enterprise teams turn messy workflows into usable AI-powered products, automations, and systems design. AI Design Engineer. Product designer who also builds: complex, workflow-heavy products taken from research through working code, with AI features grounded in real workflows — chatbots, RAG pipelines, and multi-agent systems. ## Pages - [Home](https://gilbertomedrano.com/) - [Bio](https://gilbertomedrano.com/bio) - [Work](https://gilbertomedrano.com/work) - [Publications](https://gilbertomedrano.com/publications) ## Case studies - [Workflow Builder](https://gilbertomedrano.com/work/workflow-builder): StackStorm workflows were powerful, but the logic behind them could become difficult to understand once automation expanded beyond a single action. - [EDDiE - Solar Design Tool](https://gilbertomedrano.com/work/eddie-solar-design-tool): Residential solar was hard for homeowners to understand because education, design, savings, contracts, installation, and monitoring lived across disconnected conversations and tools. - [Praedico](https://gilbertomedrano.com/work/praedico): Lab billing workflows can become difficult to trust when completed tests, site invoices, master invoices, and payment status live in disconnected views. - [HAIISS Data Migration Monitoring](https://gilbertomedrano.com/work/haiiss-data-migration-monitoring): HAIISS was a healthcare data operations product used to monitor extraction and ETL processes across multiple sites before production release. - [Consumer-Direct - Life Insurance](https://gilbertomedrano.com/work/consumer-direct): Buying life insurance online is usually intimidating. Customers face personal questions, health disclosures, state restrictions, pricing decisions, authorization documents, and uncertainty about... - [Energy Monitoring Kiosk](https://gilbertomedrano.com/work/energy-monitoring-kiosk): Commercial solar clients invested in large systems, but the value was often hidden inside technical dashboards, reports, or operational data that visitors never saw. - [Colearn Parent App](https://gilbertomedrano.com/work/colearn-parent-app): Homeschooling and enrichment planning can quickly become fragmented. Parents need to coordinate multiple learners, subjects, resources, schedules, progress updates, and safety settings without... - [Commercial Ordering System](https://gilbertomedrano.com/work/commercial-ordering-system): Gil designed Finetunio, an AI model optimization platform that unified datasets, training, evaluation, and deployment into one guided workflow for improving model performance. - [Analytics Dashboard](https://gilbertomedrano.com/work/stackstorm-analytics-dashboard): StackStorm made automation executable, but teams also needed to understand how that automation behaved over time. - [Colearn Homeroom](https://gilbertomedrano.com/work/colearn-homeroom): Homeroom teachers needed more than a roster. They needed an operational view of every family, learner, attendance signal, onboarding status, and support need across a flexible learning program. - [Colearn Homeroom Admin](https://gilbertomedrano.com/work/colearn-homeroom-admin): Flexible learning programs create operational complexity at the school level. Admins need to coordinate families, homerooms, teachers, attendance requirements, enrollment capacity, staff access,... - [Customer Support Triage System](https://gilbertomedrano.com/work/customer-support-triage-system): As Engine Yard’s platform grew, customer support teams needed a reliable system to manage incoming support requests submitted through the website. - [Exchange Bidding System](https://gilbertomedrano.com/work/exchange-bidding-system): The Exchange Bidding System needed more than quote comparison. It also needed a reliable way to manage the people and organizations participating in the exchange: internal admins, exchange agents,... - [GEO – Global Energy Optimizer](https://gilbertomedrano.com/work/geo-global-energy-optimizer): Utility-scale solar planning required teams to evaluate large sites, compare system assumptions, estimate production, and translate early feasibility into proposal-ready project data. - [Helix Calculator](https://gilbertomedrano.com/work/helix-calculator): Commercial solar design depended on engineering support, static drawings, spreadsheets, and slow back-and-forth cycles that delayed proposals and limited dealer autonomy. - [In-App Conversational Interface](https://gilbertomedrano.com/work/in-app-conversational-interface): Nudgespot needed a lightweight customer-facing widget that could live inside a client’s website and turn visitor questions into manageable conversations. - [Commercial Instant Design](https://gilbertomedrano.com/work/commercial-instant-design): Dealer-led solar proposals were slowed by fragmented design edits, unclear production impact, and handoffs between layout decisions, savings estimates, BOMs, and proposal artifacts. - [Learner Launch](https://gilbertomedrano.com/work/learner-launch): Flexible learning only works if students understand what to do next. Learners needed a calm, age-appropriate interface that translated parent plans into clear daily focus, without exposing the full... - [Managed Cloud Platform](https://gilbertomedrano.com/work/managed-cloud-platform): Engine Yard provided a managed cloud platform that allowed dev teams to deploy and operate apps without managing complex infrastructure directly. - [NextGen Biosurveillance](https://gilbertomedrano.com/work/nextgen-biosurveillance): Public health analysts needed to interpret large volumes of disease signals across regions, facilities, demographics, syndromes, and data sources without losing sight of emerging risk. - [Nudgespot Conversation Management](https://gilbertomedrano.com/work/nudgespot-conversation-management): Nudgespot needed to help growing teams manage customer conversations without scattering support, internal notes, assignments, and follow-up actions across disconnected tools. - [Product-Config - Insurance Builder](https://gilbertomedrano.com/work/product-config): Insurance products depend on long application forms, product-specific questions, state rules, eligibility logic, and reusable data fields. - [Rapid-App - Insurance Platform](https://gilbertomedrano.com/work/rapidapp): Life insurance applications were slow, error-prone, and difficult for agents to complete consistently. - [Residential Project Mgt](https://gilbertomedrano.com/work/residential-project-management): SunPower’s residential installation process relied on fragmented Salesforce screens, spreadsheets, and manual coordination across teams. - [SOL - Homeowners AI Chatbot](https://gilbertomedrano.com/work/sol-homeowners-chatbot): Homeowners had many basic solar questions, but answers were scattered across website pages, sales calls, FAQs, and follow-up conversations. - [StackStorm App](https://gilbertomedrano.com/work/stackstorm-app): Infrastructure automation was powerful, but difficult to understand, configure, and monitor. StackStorm needed a visual layer that made event-driven automation feel clear without hiding the... - [Dealer Portal](https://gilbertomedrano.com/work/dealer-portal): SunPower dealers had to manage sales, accounts, resources, orders, training, and customer follow-up across fragmented tools and scattered information sources. - [Support Monitoring Dashboard](https://gilbertomedrano.com/work/support-monitoring-dashboard): Operating a managed cloud platform required constant visibility into system health and customer incidents. - [Craftal](https://gilbertomedrano.com/work/craftal): Complex products are often planned across documents, diagrams, tickets, and conversations, leaving teams without one dependable model of how the system should work. - [Contextus](https://gilbertomedrano.com/work/contextus): Most AI tools force people to repeatedly reconstruct their goals, knowledge, preferences, and working context while hiding which information actually shaped each response. - [Finetunio - Model Optimization](https://gilbertomedrano.com/work/finetunio): Improving an AI model often means navigating disconnected scripts, datasets, training frameworks, evaluation tools, and deployment systems without a reliable way to decide what should happen first. ## AI-readable corpus Plain-markdown source documents, one topic per file: - [Index](https://gilbertomedrano.com/ai/index.md): Map of every document in this corpus. - [Profile](https://gilbertomedrano.com/ai/profile.md): Who Gil Medrano is and what he does. - [Services](https://gilbertomedrano.com/ai/services.md): What he offers and how engagements are shaped. - [Skills](https://gilbertomedrano.com/ai/skills.md): Design and engineering capabilities. - [Engagement](https://gilbertomedrano.com/ai/engagement.md): How to work with him: process, timing, next steps. - [Design process](https://gilbertomedrano.com/ai/design-process.md): How the work actually gets done. - [FAQ](https://gilbertomedrano.com/ai/faq.md): Common questions with direct answers. - [Portfolio index](https://gilbertomedrano.com/ai/portfolio-index.md): Every public case study with summary and link. Generated by Portfolio::AeoArtifactsBuilder. Do not edit by hand.