Teklu Mezgebo

Hello! I'm Teklu, a graduate of the University of Minnesota Twin Cities. I am at the beginning of my career as a software developer, and I am actively looking for my first role. Generally speaking I've enjoyed building all the different types of applications throughout my academic and personal projects, but more specifically I am interested in DevOps and Cloud Computing. I am also a certified Emergency Medical Technician, and I have experience working in clinical settings, which has given me a unique perspective on the intersection of technology and healthcare.

• Built a Blender-to-MJCF asset pipeline (STL mesh assembly, joint placement, MJCF export) converting 3D scan/CAD meshes into MuJoCo simulation models
• Built a cross-platform app (React Native / Expo, TypeScript) with multiple capture modes — plain photo and AR — plus on-device guidance for image count, resolution, and coverage to produce reconstruction-ready image sets
• Designed the capture-to-reconstruction data path to a neural reconstruction tool, mapping ARKit's camera convention to the tool's expected format to validate an end-to-end capture-to-mesh flow
• Accelerated development using Claude Code while maintaining full code ownership, reviewing all generated output, enforcing consistent architecture patterns, and optimizing prompts to minimize token usage and API cost

• Prototyped a conversational voice-agent proof of concept (ChatGPT + Twilio) that placed automated outbound calls to retrieve targeted information, testing the viability of AI telephony for the team's research goals
• Prototyped a Python data-aggregation pipeline scraping relevant market information from multiple sources into a single dataset, supporting market research and surfacing emerging trends

• Documented patient encounters in an electronic health record system for a multi-provider clinic, maintaining accuracy under continuous patient flow
• Trained new hires on clinical workflows and documentation procedures, owning onboarding for the support team

• Deployed a personal resume site to AWS using S3 for storage and CloudFront for global content delivery, securing the origin with Origin Access Control
• Configured HTTPS via an ACM certificate and connected a custom domain through Route 53, routing all traffic through CloudFront
• Built a serverless visitor counter using API Gateway, Lambda (Python/boto3), and DynamoDB, with least-privilege IAM policies scoped to specific table actions
• Automated deployment with GitHub Actions, syncing frontend changes to S3 and invalidating the CloudFront cache on every push

• Architected backend infrastructure for a HIPAA-compliant digital health platform processing FHIR R4 patient records and integrating with EHR systems
• Implemented JWT authentication with bcrypt hashing and role-based access control across 4 user roles
• Built an append-only audit logging system capturing user ID, action type, resource type, and timestamp on every PHI access event across all protected routes, modeled on HIPAA audit requirements
• Engineered a role-based caretaker access system with secure patient data delegation, enforcing HIPAA privacy compliance across multi-user clinical relationships
• Containerized the backend with and deployed to AWS ECS Fargate with a PostgreSQL database on RDS

• Coordinated sprint planning and twice-weekly standup's to align team deliverables across 3 release iterations
• Designed a composite motion system enabling arbitrary nesting of behaviors so any entity could adopt complex multi-condition movement without modifying existing classes
• Authored a unit test suite across a team codebase, achieving 90%+ fault detection across entity, motion, and physics subsystems
• Containerized the full simulation, packaging only compiled binaries and runtime dependencies to produce a portable image
• Produced user and developer documentation, including JSON scene format reference, entity and motion type guides, and UML diagrams for potential new contributors