• Built Tkyo Drift, an open-source drift monitoring tool helping developers track LLM input/output shifts across time.

• Created backend systems for dynamic embedding storage, drift metric calculations, and scalar feature logging using Node.js and PostgreSQL.

• Integrated Node.js and Python seamlessly to build high-speed nearest neighbor graphs (HNSW) for drift analysis.

• Refactored critical logic for embedding validation, error handling, and cosine/euclidean drift metrics to improve robustness.

• Worked hands-on with Transformers pipelines, PostgreSQL, Docker, and OpenAI/HuggingFace APIs.

• Focused on building scalable, modular architecture for real-time drift detection and future product extensibility.

Testing-as-a-Service for some of the world’s largest and fastest-growing consumer app providers. Via dedicated client teams, a pro network of over 15,000 testers in 100+ countries, and a comprehensive software platform, I provide a suite of flexible, scalable and on-demand testing solutions. End-to-end, usability, localization, functional, and more.