Performance, latency and cost tuning - SkilBrill

Skip to content

Home
Courses
About
Contact
Blog
- Freshers

Search for:

Home - Lessons - Google Cloud Machine Learning - M6 — Deployment, Serving and Monitoring - Performance, latency and cost tuning

Google Cloud Machine Learning Performance, latency and cost tuning

Google Cloud Machine Learning

0% Completed

M1 — Machine Learning on Google Cloud: Framing

When ML is appropriate and when it is not

Problem types: classification, regression, ranking, forecasting

Supervised, unsupervised and generative patterns

Vertex AI platform tour and workflow

Data requirements and feasibility assessment

Success metrics agreed before training

M2 — Data Preparation and Feature Engineering

Data analysis and visualisation at scale

Feature engineering patterns and transforms

Handling imbalance, leakage and skew

Vertex AI Feature Store and feature reuse

Data validation and drift detection

Serving-point versus training-point consistency

M3 — BigQuery ML

Building linear and logistic regression in SQL

Tree-based models and boosted trees

Time-series forecasting in BigQuery ML

Evaluation, explainability and feature importance

Exporting and serving BigQuery ML models

Cost and slot considerations

M4 — Vertex AI Training and AutoML

AutoML image, text, tabular and video

Custom training jobs and machine types

Hyperparameter tuning and Vertex AI Vizier

Distributed training and GPU selection

Pre-built containers and custom containers

Experiment tracking and run comparison

M5 — Custom Training and TensorFlow

TensorFlow and Keras modelling patterns

tf.data pipelines and performance

Transfer learning and pre-trained models

Model saving, versioning and export

Evaluation strategy and reproducibility

When a framework beyond TensorFlow is justified

M6 — Deployment, Serving and Monitoring

Model endpoints and traffic splitting

Batch prediction versus online serving

Feature Store online serving and freshness

Model monitoring for drift and skew

Explainable AI and attribution

Performance, latency and cost tuning

M7 — MLOps: Pipelines, Registry and Governance

Vertex AI Pipelines and orchestration

Model registry, versioning and approval

Reproducibility, metadata and lineage

Continuous training and retraining triggers

Model risk, fairness and bias evaluation

Regulatory expectations for model documentation

M8 — Capstone: End-to-End ML System

Frame a business problem as an ML task

Prepare and validate the training data

Compare BigQuery ML against a custom model

Deploy the chosen model to a monitored endpoint

Build the retraining pipeline with lineage

Present the model and its risk position

Announcements Course Info

Global Trusted Software Training Partner Since 2022

SkilBrill Training Institute delivers industry-focused technology education that bridges the gap between academic learning and real-world demands. Our structured programs combine foundational theory with practical application, ensuring every graduate is confident and job-ready from day one.

Learn More About Us

© 2026 SkilBrill Training Institute. All Rights Reserved. Privacy Policy Terms Refunds Outcomes

📞 Enroll Now

Start Your Journey

Fill in your details and our team will get back to you within 24 hours

Please enable JavaScript in your browser to complete this form.

Full Name *

Email *

Course Interested In *

Mode of Course *