{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/07800b7e1e6e4a9eb6875b93d9d4937d\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/os/loom-media-content/loom/b10ecee4-5e34-4dc1-8cfd-a91a3d5e6445/f552db6d-673c-4c76-9905-d824a770b210/sessions-thumbnails-07800b7e1e6e4a9eb6875b93d9d4937d-72a2e15ce0198065-full.jpg/binary?authToken=eyJraWQiOiJtaWNyb3MtZ3JvdXAvdGRwLW9zLWFyY2hldHlwZS9hOWI5NmI1OWlwM3JxOWRtIiwiYWxnIjoiUlMyNTYifQ.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.Gt80wiw3rYQ8fnG70XcgR6m2U3Rh0yvo41XoKU9IRZEpRrtw174vLwRFXdvklVHHkFpMELMq2KdMwK1MBuwpv4E1d5gx3tdc0MSLtZJXx-_qQ-SInSgVEqlss8C-wtgs72he5GyIIUF5eV8eY7L0gCCITvwusVrouA80j_ovsa_3xLr224E4EJWGCjRH3jJWvlwMFGnsYhtufqBW6x1LmM-6BPymIno_iocurAdxBWlXei6kkDgrp18CfyDZAbd9MQOtBzsnisUOoF1aUHuXjvvr5wkUxCQWdr37Gv-6DM6iyYMnwb6ZkZam6XBR9l2DnBLivMRvrDLU4qz1426RDg&Expires=1790982602&Signature=DhubrvyvacUsH8SJ4zMNLpWH30yHulkeEIesncqNEaQodi5z7JNFLkEJ1U7CpGp5aZPFtfvl7N1xv3ZQhmYCiJRpQy04klmUeFYiLc1zFG1-3W19dPLNrVQBPqDuHdskI5jH5syaDHE9aPnl3DIZgYBxvG3P2wIxNr3--gkF7Zc5GflWOEdxpdJHgCe2Ygq5Qimjs0o3boJJgH3j4hBwn1AuijNbNGF2G8brOe3G%7EQgE3vsPYiSw1cEUbbACXQoUO8un2Eykwf9U6eSaWJEcMB%7ExfAAioQMA6kj61WQ64YWdhtE74728DyrveE%7Ehk8yUKVjZ0L%7EbUZbLwFSTLqRGxA__&Key-Pair-Id=KQOSYIR44AIC0","duration":925.7,"title":"Terrifix Mesh - Serverless Model Fine Tuning and Inference","description":"This Loom introduces Terphix Mesh, a unified platform to fine-tune, deploy, evaluate, and govern AI across the MLOps lifecycle. It explains how the platform supports multiple teams from data and ML to development and operations, then shows an admin and user view with infrastructure health metrics including GPU memory, utilization, temperatures, tokens per second, latency, and running jobs. Terphix Mesh focuses on running open-weight models and fine-tuning jobs with an MLflow integration that automatically logs experiments and adapters, then deploying production-ready endpoints with open AI compatible APIs. The speaker notes deployment times on the order of about 3.9 minutes and describes extensive observability and alerting through system, GPU, and inference metrics plus traces and dashboards."}