Presented by

Want to appear here? Talk with us

The 2nd Edition of Continuous Cloud Usage Optimization is out.

Built by practitioners, this edition adds 15 new contributors, 100+ new real-world insights, new AI governance guidance, and expanded Snowflake and Databricks coverage. It goes beyond cost reporting to show how teams build guardrails, automate optimization, and run FinOps as an operating model.

— Published by Stacklet

It's Friday morning and you're planning an easy day when a budget alert lands in your inbox. Your manager sees it. The FinOps team sees it.

Your relaxed Friday just went ... there.

Great guide by the AWS CFM Team on how to use AWS Kiro CLI and Amazon Q Developer that can help you stop these panic moments before they start. Here are 5 ways to use these tools to cut costs and save time.

Get cost recommendations in seconds: Type "Get me compute optimizer recommendations for my account" and Kiro connects to AWS Cost Optimization Hub, pulls all your savings opportunities, and shows them right where you work.

You set this up once by adding a few lines to your Kiro settings file that connect to Cost Explorer, Pricing, and Billing tools.

Update your infrastructure files automatically: Kiro can find optimization opportunities and fix your infrastructure code for you. Tell it "Go through my infrastructure, find instances that can be optimized, change them to the recommended type, and give me the deployment command." A 15-minute task now takes two minutes.

Optimize entire infrastructure files with one click: Open any infrastructure file, select the code, right-click and choose "Optimize." Amazon Q analyzes everything and suggests improvements like switching storage from GP2 to GP3 volumes, adding S3 lifecycle policies, consolidating NAT Gateways, and migrating to Graviton instances. One customer found $2,550 in monthly savings from a single file review.

This works on application code too, finding performance and cost problems in your Python or JavaScript.

Write cost policies in plain English: Service Control Policies are powerful but hard to write. Now you can describe what you want in normal language: "Only use GP3 volumes, deny NAT Gateway creation outside the main VPC, enforce tagging on all resources." Kiro writes the complete policy JSON for you.

Before you apply it, ask Kiro to check your existing infrastructure against the new policy. It scans everything and tells you what would break, then offers to fix those violations automatically.

Teach AI your cost rules once, use them forever: Create a simple text file with your organization's standards. Write things like "Always use ARM64 architecture for Lambda, add CloudWatch logs with 7-day retention, prefer Graviton instances, use GP3 volumes by default." Save this as your context file. Now every time you ask AI to create something, it automatically follows your rules without you having to repeat them.

Read More

EC2 Capacity Manager Now reports Spot interruption metrics (total count and rate) across accounts and regions, enabling teams to quantify reliability and optimize fleet composition.

Marketplace Seller Reporting New collections visibility allows sellers to track invoice status (collected, disbursed) directly in revenue dashboards for better cash flow forecasting.

CloudTrail Lake to CloudWatch Simplified data import flow allows unified analytics. FinOps Note: While the import action is no-cost, ingested data incurs standard CloudWatch custom log charges.

Instance & Gaming Updates Graviton4 families (M8g, M8gn) and new regional expansions for C8i/R8i offer fresh price-performance options. Additionally, GameLift Streams added real-time cost metrics and warm buffer autoscaling.

Read All AWS Updates

🫙

Read All Azure Updates

Future Resource Availability Now Generally Available, you can view future availability for GPU, H4D, or TPU VMs in calendar mode before creating a reservation request. This visibility helps increase approval likelihood for high-demand resources.

Read All GCP Updates

​Join FinOps Weekly Call: 2026 Year Launch. Here’s what we are going to do:

​Bring your favorite drink and one cost challenge or clever hack to share — there’ll be time for rapid-fire wins and open mic troubleshooting.

​Whether you want to pick up new ideas, show off a success, or help shape the call plans for the year, this session is made for getting things moving (and having a little fun while we do it).

This guide shows you how to set up automatic cost alerts that run entirely inside Snowflake using just SQL and Slack webhooks. The system sends a weekly report to your team every Monday morning showing daily credit usage and costs for the past seven days.

The system calculates your seven-day average spend and marks any day that goes over budget with a red alert emoji. Days under budget get a green checkmark.

First, you create a Slack webhook that connects to a dedicated channel for cost alerts. Then you set up a Snowflake notification integration that can send messages through that webhook.

Next, you build a SQL procedure that pulls daily usage data from Snowflake's built-in metering tables and formats it into a readable report. Finally, you create a scheduled task that runs every Monday at 10 AM to generate the report and send it to Slack.

The procedure pulls your actual contract rate from Snowflake's rate sheet tables to calculate real dollar costs, not just credit usage. If your account doesn't have access to those tables, you can hardcode your rate as a fallback.

Read More

Join us as we talk with Olfa from Azure Cloud Academy about why FinOps implementations often take longer than expected. We explore the gaps between FinOps theory and real-world production, the importance of understanding business value, and the role of cultural practices in achieving cloud cost efficiency.

Listen on Apple Podcast - Listen on Spotify - Watch on YT

Your AI project just passed every technical test, but now it's stuck in a review process that will cost you three more months and hundreds of thousands in developer time. This is the reality most companies face when they try to put AI into production.

The problem isn't the technology. It's that three different teams are fighting over the same project with completely different goals.

Your FinOps team wants to cut cloud costs. Your risk and compliance team wants to avoid lawsuits. Your AI team just wants to ship the product. Nobody is talking to each other, and the project dies in the middle.

Here's what most companies miss: AI governance isn't free. It comes with massive hidden costs that never show up on your cloud bill.

Before you even launch, you'll pay for development rework. You'll also pay for documentation. Once your model goes live, the costs get worse. You need constant monitoring for bias and fairness, not just accuracy. You must store every prediction and input for years to meet audit requirements. That storage bill grows every single day.

The gap between regulated and non-regulated AI is huge. A social media app that recommends the wrong video costs nothing. A bank that denies mortgages unfairly faces federal investigations and million-dollar fines.

Read More

AWS isn't dying, but 2026 will test whether they can still run things as well as they used to. That's the core message from Corey Quinn's analysis of where AWS stands heading into this year.

The company is printing $132 billion annually with 29% market share, so they don't need anyone's sympathy. But being "fine" isn't the same as being "dominant the way they used to be." Three big shifts happened that AWS resisted for years:

They now support multi cloud with AWS Interconnect playing nice with Google Cloud and Azure.

They made serious on premises investments with AI Factories, responding to reports of 21% workload repatriation and 86% of CIOs planning to bring something back on prem.

They democratized model training with Nova Forge at $100,000 annually, making custom models accessible to companies that merely have money instead of "build a moon base" money.

AWS made the right strategic moves and finally accepted market reality instead of insisting the market was wrong. What's less certain is whether they can retain the people who know how to deliver on those strategies when everything's on fire at 3 AM.

Read More

AI FinOps Expertise

Vaibhav provided wonderful insights during our AI webinar, analyzing how to optimize and control LLM costs. A wonderful member of our community!

Connect with the Professional of the Week

Azure FinOps Cost Optimization Course
(Starting February 2026)

Enrollment Opened

Alfonso is creating a masterpiece.

The biggest educational we’ve done till date.

Learn all you need to optimize resources in Azure.

199€ Super Early Bird – Ends December 21st (CLOSED)

⏭ 299€ Early Bird: Opens January 2nd – Ends Tomorrow

499€ Launch Offer: Opens XXX

699€ Starting Offer: Opens XXX

999€ Final Price