Blocks ads

If you use AWS, this will hurt. 20-40% of your AWS spend is wasted. And deep down, your team already knows it. Not because they’re bad at their jobs. Not because your setup is uniquely broken. And not because nobody has looked at it before. It’s because AWS cost optimization sits in that dangerous category of problems every scale-up has: expensive enough to matter, but never urgent enough to win against product deadlines, customer requests, hiring, fundraising, or growth targets. So the waste stays. The inefficiencies stay. The surprise bills stay. And your margin quietly takes the hit every single month. Blocks fixes that. We guarantee at least 20% AWS savings.. No upfront cost. No setup fees. No long-term lock-in. No infrastructure changes.

If you use AWS, this will hurt. 20-40% of your AWS spend is wasted. And deep down, your team already knows it. Not because they’re bad at their jobs. Not because your setup is uniquely broken. And not because nobody has looked at it before. It’s because AWS cost optimization sits in that dangerous category of problems every scale-up has: expensive enough to matter, but never urgent enough to win against product deadlines, customer requests, hiring, fundraising, or growth targets. So the waste stays. The inefficiencies stay. The surprise bills stay. And your margin quietly takes the hit every single month. Blocks fixes that. We guarantee at least 20% AWS savings.. No upfront cost. No setup fees. No long-term lock-in. No infrastructure changes.

Here's exactly how I talked to a CTO out of two hires he didn't need. He thought I was going to help him build the case for headcount. Two open reqs sitting there for months. His best engineers stuck with firefighting instead of building. Two more hires, he figured, and it would sort itself out. We didn't start there. We mapped his OPS first. This is how I always start: Someone comes to me with a headcount problem, a retention problem, a "my team can't keep up" problem. I don't touch any of that yet. I sit with the team and map where the hours actually go. Not what they THINK happens. What ACTUALLY happens, hour by hour, for a real week. Sat down and audited where the team's time actually went. Meetings, handoffs, decisions nobody owned, setup work that ate a week before anyone touched the real product. After enough of these audits, the patter is same: The company changes. The industry changes. The team size changes. What breaks doesn't. It's almost always the same three things. → Decisions with no clear owner, so everything routes back to one person → Setup and maintenance work nobody budgeted time for, so it eats the roadmap quietly → A cost or risk sitting untouched because nobody was ever assigned to watch it This CTO had all three. But the noisiest one was the third. Then I asked him the question that actually matters before you hire anyone. What do you need this person to do, exactly? He wasnt sure about that one ... So we mapped it down. Task by task. Hour by hour. What the hire would actually spend their week on. Most of it turned out to be the same thing over and over. Chasing cost alerts. Chasing security flags. Cleaning up stuff nobody had time to fix properly. A task can get solved without a new seat. So I asked him something else.: Is there a tool for this already? I don't know that world. I'm not a technical person. But he went and looked, and he found Blocks it runs an agentic DevOps engineer called Major Tom straight on the AWS account. It scans for waste and security issues and hands the team a ready Terraform PR to fix it. Someone reviews it, someone merges it. No new hire needed to babysit . Within weeks: → 27 cost findings found and fixed on one account → Zero new hires needed → His engineers back on the roadmap instead of the fire Seeing the mess and fixing it, without adding a seat, thats a solution. Every company thinks their mess is unique. It rarely is. It's almost always ownership, maintenance, or visibility. Find which one and you'll know exactly what to fix, and whether a tool can fix it before you post a single job. CTO. CEO. What keeps breaking and costing you in your ops? If your ops are messy and costly and you don't know where to start → let's talk. Paid partnership with @Blocks, check if they can help with your costs: https://lnkd.in/d-YvCVjP Alisa Reznik : fCOO & Executive Adviser : I build leaders who build companies.

Your team knows its cost per token to three decimal places but can't tell you how many NAT gateways it runs. The second number matters more than you think: 🚨 NAT gateway, idle: $394 a year, before a single byte 🚨 NAT data: 1 TB a month, $540 a year 🚨 gp2 vs gp3 volumes: $0.10 vs $0.08 per GB-month 🚨 EBS snapshots: 5 TB of old ones, $3,000 a year Sources: aws.amazon.com/vpc/pricing and aws.amazon.com/ebs/pricing, us-east-1, September 28. The token bill has a dashboard, a budget, and a weekly review. The NAT gateway created for a staging VPC by an engineer who left in 2024 has none of those. It bills $394 per gateway per year and never pages anyone. The gp2-to-gp3 migration saves $240 per terabyte per year, and the ticket is never written. The waste is small, plural, and permanent. No single line justifies an engineer-day. Together they are the cleanup sprint that keeps slipping because the model has to ship. Blocks puts this bucket at 5-15% of the bill. That is what Major Tom, the AI agent Blocks runs against your account, is for. It scans daily, flags waste in Slack, and opens a Terraform PR for each fix. You review and merge. A DevOps lead quoted by Blocks: "Major Tom caught an RDS Extended Support charge I'd completely missed." Major Tom is one of three levers. Group buying and managed Savings Plans are the other two, and Blocks guarantees 20% off the AWS bill. No savings, no fee. A platform team with real cycles can run this audit itself, and should! When did anyone on your team last list the NAT gateways? 👉 Check out their website: https://lnkd.in/eQJJPtds

Read the Blocks report on AWS savings they keep finding across 100+ startups. Filtered out anything under $200/yr and still landed on 10 patterns that repeat everywhere. The two I see constantly, in basically every account I touch: • NAT gateway sprawl: one per AZ "for redundancy," half of them idle, ~$32/mo each just to exist before data transfer even kicks in • Cross-AZ traffic nobody accounted for when they split subnets Their numbers back it up: - EC2 rightsizing hits 46% of accounts, median $673/yr (!!) - RDS rightsizing is rarer at 14% but the biggest single line item, $3,504/yr median 😅 🔥 - And manual RDS snapshots don't expire like automated ones do, so they just sit there billing you forever. btw: it's not a visibility problem. People know the NAT gateway is idle. Nobody prioritizes killing it 🥲 Worth a read: https://lnkd.in/eCC5ZJJT Great article by Dr. Andreas Schroeter & Rijul Gogia 🙏

Cloud infrastructure has become easier to provision, but keeping it cost-efficient and secure is still surprisingly manual. The harder problem is often not identifying an issue, but turning that finding into a safe, reviewable change without adding more operational overhead. I find this shift toward agentic DevOps particularly interesting: moving from dashboards that tell us what is wrong to agents that can actually help us fix it. • Continuously scan cloud infrastructure for cost waste and security misconfigurations. • Turn findings into concrete fixes rather than another alert or ticket for someone to investigate. • Generate Terraform changes that fit existing infrastructure-as-code workflows and remain human-reviewed. • Meet engineers where they already work, including Slack, GitHub, AWS, and MCP-enabled development environments. Blocks is taking this approach with Major Tom, its agentic DevOps engineer, which scans AWS environments and can turn findings into Terraform PRs for teams to review and merge. More on Blocks: https://lnkd.in/gvfvDcHf