Kevin Bravo

Free2Z is a creator platform for publishing content, livestreaming, and receiving direct community support. I work across the product stack, from user-facing flows to backend systems and production infrastructure.

• — Led the end-to-end migration of the legacy React frontend to SvelteKit, reaching full feature parity while the platform remained in production.
• — Improved the mobile and desktop experience, including responsive layouts, navigation, and core publishing flows.
• — Improved the article creation flow and added platform internationalization, including multilingual article authoring.
• — Build and maintain full-stack features with SvelteKit and TypeScript on the frontend and Python, Django, and PostgreSQL on the backend, including REST APIs, authentication, livestreaming, and content systems.
• — Hardened Zcash challenge authentication against replay and proxy-spoofing attacks by persisting single-use validations in PostgreSQL and updating the OpenAPI contract, NGINX policy, deployment configuration, tests, and runbooks.
• — Currently developing a Model Context Protocol (MCP) server and strengthening backend and infrastructure components for better security, reliability, and maintainability.

• Product Engineering
• SvelteKit
• TypeScript
• Python
• Django
• PostgreSQL

Billingua Talent was a multi-tenant talent and program management platform commissioned by the U.S. Department of State’s Venezuela Affairs Unit and developed through AVAA. It was designed to connect bilingual Venezuelan professionals with companies hiring in Venezuela and give advisors tools to manage candidates, training services, and program data.

• — Owned end-to-end technical and product delivery as the sole engineer, leading requirements discovery, software architecture, data modeling, product design, full-stack implementation, testing, and launch readiness.
• — Worked directly with non-technical stakeholders at AVAA and the U.S. Embassy, translating program needs and feedback into requirements, product flows, and shipped features while explaining technical decisions and trade-offs in accessible language.
• — Designed the multi-tenant architecture and role-based access control for candidates, companies, advisors, and administrators.
• — Built candidate profiles, job opportunity flows, English-level tracking, course and service management, registrations, dashboards, analytics, and reporting.
• — Built the bilingual interface and the frontend and backend systems supporting the main user and administrative workflows.
• — Prepared the complete application for launch and coordinated the first testing sessions with AVAA and the embassy team, validating core flows and incorporating their feedback.

• Product Development
• Next.js
• Multi-tenancy
• RBAC
• Analytics & Reporting

Independent software engineer for clients in education, finance, and professional services. I worked directly with each client to turn loosely defined needs into websites, mobile applications, automations, and full-stack features.

• — Owned each project end to end as the sole engineer, including requirements discovery, scope definition, technical and product decisions, implementation, deployment, iteration, and post-launch support.
• — Designed, built, and launched Astro websites, including jenndolinski.com and rodolfomedinadelrio.com, owning responsive UI, content structure, forms, performance, and deployment.
• — Built Proyecto Educa’s website as a freelance engagement; the organization later invited me to join as Chief Technology Officer.
• — Developed cross-platform mobile applications with Flutter and Dart.
• — Created automation scripts and full-stack features around existing client workflows, working across frontend, backend, integrations, and data cleanup as each project required.
• — Worked directly with non-technical clients, communicating progress, explaining technical decisions in accessible language, and incorporating feedback through short iteration cycles.

• Web Development
• Product Engineering
• Astro
• Flutter
• Dart
• Automation

AVAA and EducationUSA managed core operations through disconnected spreadsheets. As the sole engineer, I turned those processes into two production platforms and owned their delivery end to end: product discovery, architecture, data modeling, full-stack development, Azure deployment, and support.

• — Built SEP, the Participant Evaluation System, with Next.js, PostgreSQL, Prisma, and Auth.js. It became ProExcelencia’s primary platform across all three states where the program operates, now serving around 400 monthly active users and centralizing scholar records, activities, attendance, volunteer hours, and reporting.
• — Built SEA, the Advisee Evaluation System, for EducationUSA operations, centralizing student histories, outreach activities, services, payments, receipts, institutional partnerships, and reporting.
• — Worked directly with non-technical staff and scholars to understand their processes, present progress, and turn requirements and feedback into product flows and features.
• — Designed the data models, permissions, and workflows for both platforms, deployed and operated them on Microsoft Azure, monitored production, resolved issues, and shipped updates.
• — Implemented a staff-only retrieval-augmented generation feature in SEP that retrieves record context before generating answers, helping staff query program data and surface insights without manually searching records and reports.
• — Helped establish SEP’s maintenance program, through which AVAA scholars maintain the platform, resolve issues, and develop features after its initial implementation.

• Full-Stack Development
• Data Modeling
• Next.js
• PostgreSQL
• Microsoft Azure
• RAG

ProExcelencia depended on manual processes and scattered spreadsheets. As a software engineering intern, I consolidated four years of historical data, automated recurring workflows, and built the initial database for SEP, the internal platform I later developed as an AVAA engineer.

• — Created an interactive Power BI dashboard covering scholar demographics, academic records, employment, satisfaction, and outcomes, giving the team its first consolidated view of four years of program history.
• — Developed Python ETL scripts to clean, normalize, validate, and reconcile inconsistent records from multiple spreadsheets, creating a reliable dataset for analytics and reporting.
• — Automated recurring activity creation and management with Google Apps Script, reducing manual work for program staff.
• — Recovered and structured scholar activity data from 2020 to 2023. This became SEP’s initial dataset and helped the platform replace spreadsheet-based operations across all three states where ProExcelencia operates.

• Microsoft Power BI
• Business Intelligence
• Python ETL
• Google Apps Script
• Data Cleaning