Uploaded July 2026 | Updated September 2026, 3 weeks ago
GitHub repo → https://goo.gle/4y6rlVn
Agent Platform observability → https://goo.gle/4eEgzgd
Google Antigravity → https://goo.gle/3R5gUki
Cloud Run → https://goo.gle/4f3VY6p
In this action packed episode of the AI Agent Clinic, developer Sami Maghnaoui brings his sports app, Playback IQ, into the clinic to tackle a major challenge: sluggish generation times during a global football tournament! ⚽️
Watch as AI expert Luis Sala and Sami fire up the Antigravity IDE to diagnose hidden performance bottlenecks. See firsthand how they use Google Cloud’s Gemini Enterprise Agent Platform—the premier unified environment to build, deploy, govern, and optimize enterprise grade AI agents—to completely transform their workflow.
Discover how a brilliant mix of OpenTelemetry instrumentation, parallel processing, and a swift Cloud Run deployment took a slow build and turned it into a lightning fast, production ready agent. 🚀
💡 Crucial GenAI Cost Hack: Learn how to track token consumption directly inside your telemetry tools! Watch along and learn how to monitor these metrics to continuously maintain high quality outputs once an app goes live.
Do you want to be a guest on The Agent Clinic and have your agent fixed live? Submit your repo to: agent-clinic@google.com.
Watch along and learn how to:
* Optimize on the agent platform: Use the platform's robust capabilities to slash processing time by 80% by handling SDK outputs and executing voice synthesis in parallel.
* Monitor tokens & cut costs: Integrate token consumption tracking to set the stage for lowering costs
* Pinpoint latency with Antigravity: Use OpenTelemetry standards inside the Antigravity IDE to send data to Google Cloud Trace, breaking down execution into visual "spans."
* Deploy seamlessly: Take your optimized agent straight to production using Cloud Run.
Chapters:
00:00 - AI agent latency issues in production (PlaybackIQ case study)
05:30 - Architecture overview: Gemini agent + Vector DB for live sports data
12:15 - Diagnosing high LLM response times under concurrent traffic
18:40 - Configuring OpenTelemetry Tracing for Gemini Enterprise agents
🔔 Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech
#AIAgentClinic #GoogleCloud
Speakers: Luis Sala, Sami Maghnaoui
Products Mentioned: Gemini, Antigravity, OpenTelemetry, Cloud Run, Cloud Trace
GitHub repo → https://goo.gle/4y6rlVn
Agent Platform observability → https://goo.gle/4eEgzgd
Google Antigravity → https://goo.gle/3R5gUki
Cloud Run → https://goo.gle/4f3VY6p
In this action packed episode of the AI Agent Clinic, developer Sami Maghnaoui brings his sports app, Playback IQ, into the clinic to tackle a major challenge: sluggish generation times during a global football tournament! ⚽️
Watch as AI expert Luis Sala and Sami fire up the Antigravity IDE to diagnose hidden performance bottlenecks. See firsthand how they use Google Cloud’s Gemini Enterprise Agent Platform—the premier unified environment to build, deploy, govern, and optimize enterprise grade AI agents—to completely transform their workflow.
Discover how a brilliant mix of OpenTelemetry instrumentation, parallel processing, and a swift Cloud Run deployment took a slow build and turned it into a lightning fast, production ready agent. 🚀
💡 Crucial GenAI Cost Hack: Learn how to track token consumption directly inside your telemetry tools! Watch along and learn how to monitor these metrics to continuously maintain high quality outputs once an app goes live.
Do you want to be a guest on The Agent Clinic and have your agent fixed live? Submit your repo to: agent-clinic@google.com.
Watch along and learn how to:
* Optimize on the agent platform: Use the platform's robust capabilities to slash processing time by 80% by handling SDK outputs and executing voice synthesis in parallel.
* Monitor tokens & cut costs: Integrate token consumption tracking to set the stage for lowering costs
* Pinpoint latency with Antigravity: Use OpenTelemetry standards inside the Antigravity IDE to send data to Google Cloud Trace, breaking down execution into visual "spans."
* Deploy seamlessly: Take your optimized agent straight to production using Cloud Run.
Chapters:
00:00 - AI agent latency issues in production (PlaybackIQ case study)
05:30 - Architecture overview: Gemini agent + Vector DB for live sports data
12:15 - Diagnosing high LLM response times under concurrent traffic
18:40 - Configuring OpenTelemetry Tracing for Gemini Enterprise agents
🔔 Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech
#AIAgentClinic #GoogleCloud
Speakers: Luis Sala, Sami Maghnaoui
Products Mentioned: Gemini, Antigravity, OpenTelemetry, Cloud Run, Cloud Trace
