Hello, my name is
Alperen Ugus
Senior Software Engineer
Distributed Systems • Full-Stack • Agentic AI
About Me
Senior Software Engineer with over 5 years of experience architecting and scaling resilient distributed systems within the healthcare and fintech industries. Proven track record of leading large-scale cloud migrations and building high-performance full-stack applications that drive operational efficiency. Expert in integrating Agentic AI to solve complex automation challenges and dedicated to engineering excellence through robust observability and secure, scalable architecture.
Years Experience
5+
AWS Certified
Dev Associate
Education
Colorado School of Mines, CS Cum Laude
Experience
Sr. Software Engineer
eHealth, Inc.
Apr 2025 – Present
Austin, TX
- Architected a centralized appointment reminder system using Java, AWS SQS, and Redis, improving appointment attendance rates by 30%.
- Implemented a lazy loading strategy for eCommerce platforms, reducing initial page load times by 15-20%.
- Mentored the engineering team on microservice architecture, request lifecycle management, and GraphQL implementation.
Sr. Full Stack Software Engineer
TriTech Software
Aug 2023 – Apr 2025
Dallas, TX
- Led the migration of a legacy desktop application to a cloud-based web application with microservice architecture and function-level security.
- Designed the Stratus UI with React/Redux and established a reusable component library using Storybook and Material UI.
- Architected an admin application automating product access management, reducing operational overhead by 95%.
Sr. Associate, Software Engineering
Capital One
Feb 2022 – Aug 2023
Dallas, TX
- Redesigned high-traffic platforms through A/B testing and modular architecture, optimizing the journey for 300k+ daily visitors.
- Architected a real-time data synchronization system using event-driven patterns, eliminating daily database batch syncing.
- Engineered an enterprise observability framework for real-time monitoring and proactive issue resolution.
Software Engineer
eHealth, Inc.
Oct 2021 – Feb 2022
Santa Clara, CA
- Developed full-stack features improving user engagement and implemented centralized observability documentation.
- Reduced troubleshooting and support resolution times by 50% through integration documentation and monitoring.
Software Engineer
Zayo Group
Feb 2021 – Oct 2021
Boulder, CO
- Automated frontend development workflows, reducing engineering time for new components by 90%.
- Optimized application performance through advanced caching strategies, achieving a 40% reduction in response times.
Skills & Technologies
Languages
Frontend
Backend
Cloud & DevOps
Agentic AI
Featured Projects
MasalAmca
AI-powered Turkish bedtime stories for children — personalized narration and minimal screen time. Native SwiftUI app (iOS 17+, iPhone & iPad) generates safe, unique stories per child (name, age, themes), narrates them with natural AI voices, and keeps listening friction-free with background playback, lock-screen controls, optional white noise, and a saved-story library. Shipped with SwiftData, OpenAI GPT-4o-mini, Gemini Flash TTS, a Cloudflare Worker proxy for auth, rate limits, and server-owned prompts, plus RevenueCat and StoreKit 2 — a production App Store product focused on imagination-first, screen-optional listening.
Lullaby: Baby & Bump Tracker
iOS companion for pregnancy through early childhood: feeds, sleep, diapers, health notes, growth, and milestones. Caregivers log in seconds from the dashboard; data stays on-device and syncs through private iCloud (SwiftData + CloudKit) — no custom backend or third-party trackers. Soothing lullaby playback with background audio and lock-screen controls for calm routines. Native SwiftUI (iOS 17+, watchOS, widgets, App Intents), strict Swift 6 concurrency, accessibility-minded UI, and RevenueCat for Lullaby Plus — a production App Store product for privacy-sensitive families.
RAG4PDF
Production SaaS for cited, grounded answers from your own PDFs via a multi-agent RAG pipeline. Users sign in with Google, configure AI agents (each with its PDF library and optional persona), and chat with responses that reference exact book and page; sources open the PDF at the cited page in the browser. Features parallel hybrid retrieval (vector + full-text), query expansion, passage reranking, NLI-based grounding checks, and async answer evaluation per agent; multi-stage ingestion with background jobs into PostgreSQL and pgvector; Stripe billing with idempotent webhooks and a local mock path; Google OAuth with JWT and shared-agent demo links. Spring Boot 3.5 / Java 21 backend with LangChain4j and OpenAI, React / TypeScript frontend with SSE for ingestion status and a full billing UI — deployed on Railway with Docker Compose for local development.
Certifications
Get In Touch
Feel free to reach out. Whether you have a question
or just want to say hi, my inbox is always open.