Ajit developed foundational telemetry infrastructure for the huggingface/xet-core repository, focusing on observability and reliability. He implemented a buffering layer and dedicated logging path to capture and deliver telemetry data, ensuring robust data flow with staging fallbacks during deployments. Using Rust and Python, Ajit extended client-side telemetry by integrating repo_id extraction and logging within core modules, enabling usage analytics and faster issue triage. He also stabilized serialization dependencies by updating serde_json usage, reducing build and runtime issues. Ajit’s work demonstrated depth in system design, data engineering, and dependency management, laying groundwork for future analytics and improved developer efficiency.