Toronto, Ontario, Canada
Software Engineer
I build production software across web, cloud, AI, and iOS: an App Store app, a real-time voice AI product, and a full-stack AWS platform.
Three products. Where I can share the engineering, the case study covers the architecture, my role, and the decisions behind it. The repositories are private, so no source code is exposed.
Independent full-stack AI / AWS platform
A spoken mock-interview companion that gives warm, usable suggestions instead of scores, built full-stack on AWS with asynchronous AI analysis and per-tenant data isolation enforced by IAM.
Independent project. I built it end to end: the Next.js app and server APIs, AWS processing, authentication and authorization, CI/CD, and tests.
348passing automated tests
Real-time voice AI learning product
A browser-based dance tutor that teaches K-pop choreography with pre-recorded instructional clips and a real-time voice AI coach.
Co-built with a co-founder. My focus: the SessionController state architecture, LLM tool handlers, ElevenLabs agent and tool configuration, frontend flows, and the AWS telemetry pipeline.
~100testing sessions before the product pivot
Native iOS product
A native iOS dance-practice app: users record or import video, then review and practice dance content.
Co-founder, September 2025 – June 2026. I am now a former co-founder.
What I have actually built, and where to see it. Each area links to the case studies that show it.
State-heavy client flows in React and TypeScript, and an authenticated Next.js application with server APIs.
Built asynchronous AWS workflows, APIs and multi-tenant data systems, and delivered static sites through S3 and CloudFront.
Authentication with Clerk, defense-in-depth tenant isolation enforced through IAM, and CI that authenticates to AWS with OIDC instead of stored keys.
LLM tool calling with deterministic state enforcement, structured outputs validated against schemas with retry and fallback, and real-time voice.
A concurrency-safe AVPlayer engine, actor-based local-first storage with atomic writes and schema migration, and coordinator-based navigation.
Seen in: Boog
CI/CD pipelines in GitHub Actions on all three products, an automated test suite on Clovor, and crash, performance and error monitoring.
I'm a software engineer based in Toronto. With a co-founder, I built and shipped Boog, a native iOS app on the App Store, and ZeroPoint, a browser-based voice AI product that ran about 100 testing sessions before pivoting. On my own, I built Clovor, a full-stack AI platform on AWS.
Across these projects, the recurring work has been the parts that have to stay correct in production: concurrency-safe playback and storage, deterministic state around unpredictable LLM output, tenant isolation enforced at the data layer, and CI/CD that authenticates without long-lived credentials.
I'm looking for software engineering roles across full-stack, backend, frontend, cloud, AI application and iOS work.
Certification · 2026
Education · December 2022
I'm open to software engineering roles. Email or LinkedIn is the quickest way to reach me.