resume

Core Backend and Real-Time Transcription Teams

• Built and maintained Python services using aiohttp for emergency-response location and real-time transcription workflows, integrating WebSocket audio, S3 storage, speech providers, and downstream APIs.
• Built multilingual speech workflows with model evaluation, confidence controls, provider fallback, audio transcoding, and speaker-language assignment.
• Built a reusable evaluation tool for comparing speech-to-text models using live and prerecorded audio before production rollout.
• Kept transcripts available during audio-service outages and propagated call IDs across services to improve traceability and incident debugging.
• Preserved audio continuity across WebSocket reconnects and Kubernetes rollouts by resuming S3 uploads, tracking active sessions, and implementing graceful shutdown.
• Modernized services to Python 3.10, replaced legacy XML parsing, added Datadog metrics and dashboards, and resolved production incidents in a PagerDuty on-call rotation.

Web Team

• Built subscription upgrades across Google Play, the Apple App Store, and Stripe, increasing subscription upsells by 40% and reducing upgrade-related support tickets by 80%.
• Restructured the user-folder data model and Django API, reducing folder response times by 55%.
• Implemented cross-product authentication and subscription-benefit access between Outside and Gaia GPS following the acquisition.
• Refactored the sync API from atomic to non-atomic transactions, reducing failed sync attempts for iOS and Android clients.

• Verified natural-gas pipeline designs and maintained documented traceability from design through construction.
• Trained and led engineers performing quality control, production tracking, and data-driven planning across design and construction teams.