Guido Asbun

• Developed and shipped production front-end features in React, Next.js, and TypeScript, reducing page load times by 15%
• Authored and maintained unit tests using Jest under a test-driven development workflow, improving code coverage and reliability
• Built and maintained CI/CD pipelines using Docker, Jenkins, and CircleCI, enabling continuous integration across the team
• Collaborated cross-functionally with engineers, product, and QA using Git, Jira, and Agile/Scrum methodologies including daily stand-ups and sprint planning
• Produced technical documentation for frontend components and participated in code reviews

• Served as the sole engineer on a cross-functional product team, building a full-stack React Native and Firebase mobile prototype for a non-profit partner and delivering 14 screens and 5 reusable components for the end-of-program demonstration
• Integrated geolocation APIs for real-time mapping features and built the Firebase data layer alongside an assigned engineering mentor
• Partnered with mentors on a scope review that cut an AR camera feature from the build, protecting the delivery timeline for the core product

Springfield, VA and Laguna, CA

Progressed from Package Loader to Driver, Pre-Load Supervisor, Driver Supervisor, and Dispatch Supervisor

• Supervised dispatch for an entire package center, planning and adjusting 55 to 60 delivery routes at a time to protect daily service commitments
• Anticipated the iPhone X launch surge at the customer counter and redesigned the will call workflow, adding an alphabetical staging layer over the existing city-based cage layout
• Cut customer wait time from over an hour to under 20 minutes and held that standard every night of launch week, with zero lost high-value packages and full security compliance; the process was reused for later launches
• Directed dispatch during a snowstorm, releasing drivers early to protect air delivery commitments and recovering ground volume through three pre-planned shuttle meet points coordinated over DIAD handhelds
• Led a same-day recovery of several hundred expedited packages after an end-of-day trailer check missed loaded trailers, then implemented a two-supervisor yard verification procedure to remove the single point of failure from the close-out process

GameShift Live — Real-Time Multi-Sport Telemetry Platform

Java 21, Spring Boot, Next.js 16, AWS Kinesis, Lambda, Bedrock, DynamoDB, ECS Fargate, Terraform

• Built and deployed an event-driven live sports platform at dev.gameshift.live that polls API-Sports, normalizes six sports into a unified SportEvent model, and publishes to Kinesis Data Streams from a Spring Boot producer on ECS Fargate
• Implemented a Java 21 Lambda Kinesis consumer that timestamps events, writes to DynamoDB with 7-day TTL, and invokes Amazon Bedrock (Claude Haiku 4.5) with sport-specific prompt templates to generate live AI color commentary
• Streamed events and commentary to a Next.js 16 dashboard over Server-Sent Events with a client-side time-travel buffer that syncs the feed to the viewer broadcast delay, plus AWS Polly text-to-speech narration of live commentary
• Enforced least-privilege IAM per service, Cognito JWT validation with Google federated sign-in, and runtime Secrets Manager injection so no credentials or API keys live in code or images
• Provisioned VPC, ALB, ECS, Kinesis, Lambda, DynamoDB, Cognito, ECR, and CloudWatch monitoring through modular Terraform with S3 remote state, deployed by GitHub Actions OIDC building ARM64 images to ECR with property-based test suites (jqwik, fast-check) gating the pipeline

Recipe AI Finder v2

Java 21, Spring Boot, Next.js, AWS Bedrock, DynamoDB, S3, ECS Fargate, Terraform

• Deployed full-stack production application at recipe-ai-finder.com using a Java/Spring Boot backend and Next.js frontend, containerized and running on AWS ECS Fargate ARM64
• Integrated AWS Bedrock for multi-model AI inference, supporting Claude, Amazon Nova, and Meta Llama 3.1 with per-request model selection and tailored prompt formats for each model family
• Implemented least-privilege IAM, GitHub Actions OIDC authentication (no static credentials), and runtime Secrets Manager injection for API keys, keeping no secrets in code or version control
• Designed modular Terraform infrastructure (VPC, ALB, ECS, DynamoDB, S3, Cognito, ECR, IAM) with S3 remote state and DynamoDB locking for concurrent-safe deployments
• Built CI/CD pipeline using GitHub Actions with Docker Buildx (linux/arm64) targeting ECR and ECS, deploying tagged images per environment (dev/prod)

Chat-Sec v2 — Secure Encrypted Messaging

C#, ASP.NET Core, TypeScript, Next.js, SignalR, AES-256, RSA-4096, Docker

• Designed and built a secure real-time messaging system with end-to-end encryption using AES-256 symmetric and RSA-4096 asymmetric key exchange, deployed at chat-secure.com
• Implemented digital signature verification for message integrity and authenticity, with secure WebSocket communication via SignalR
• Containerized the full application with Docker and validated the security implementation through targeted security testing

iOS Recipe Generator

SwiftUI, MVVM, Combine, Swift async/await, OpenAI API

• Built a native iOS recipe generation app using SwiftUI with MVVM architecture and OpenAI API integration for AI-powered recipe suggestions
• Implemented reactive data flow using Combine and Swift async/await concurrency for a smooth, responsive user experience