Understanding AI performance and cost management with New Relic
summary
Five-minute overview of using New Relic to spot LLM bottlenecks and cut costs with concrete steps: capture end-to-end traces across app, LLM and DB, inspect per-request timelines and errors, and view per-request token usage to decide between pay-for-LLM or training in-house. Recommends RAG and regular prune dataset cycles to reduce latency and hallucinations. Demo shows a sample chat app dashboard with request lists, expanded spans, error details and linked logs for rapid root-cause and cost optimization.
AI-written summary · about
related talks
48mHands-on guide to monitor your API-driven AI/LLM applications
24mTransforming Observability: Inside New Relic's AI Assistant
38mAI Monitoring Mastery: Optimize Performance, Ensure Compliance, and Cut Costs
46mLeveraging AI and Automation in DevOps
28mMastering AI Issue Debugging: A Data Scientist's Guide
16mWhat Did the AI Just Do?: Making LLMs Observable with Open Telemetry
40mOptimizing AI/ML Workloads
29mElevating generative AI and LLM observability