iiinova — where your digital self keeps evolving

Matteo Depascale @matteodepascale.bsky.social · 02/03/2026

If you've ever had to delete and recreate an entire Cognito app client just to rotate a secret, you know how overdue this was. #AWS #CloudSecurity 4/4

000

Matteo Depascale @matteodepascale.bsky.social · 02/03/2026

Why this matters: → Custom secrets = easier migration from other auth systems → On-demand rotation via AddUserPoolClientSecret API → Fits actual compliance and security requirements Not just a nice-to-have — this was a real gap. 3/4

100

Matteo Depascale @matteodepascale.bsky.social · 02/03/2026

How Cognito secret rotation works now: 1. Add a second secret (up to 2 active per client) 2. Update your apps to use the new one 3. Delete the old secret No downtime. No big bang credential swaps. Works via Console, CLI, SDKs, and CloudFormation. 2/4

100

Matteo Depascale @matteodepascale.bsky.social · 02/03/2026

Amazon Cognito now lets you bring your own client secrets and rotate them without downtime. Before: one secret, no rotation, delete the whole client to change it. Now: proper lifecycle management. Finally. 1/4

110

Matteo Depascale @matteodepascale.bsky.social · 27/02/2026

What it monitors: • Kernel issues & process limits • VPC CNI/network failures • Storage I/O & throughput limits • Container runtime failures • GPU/accelerator errors (NVIDIA, AWS Neuron) All open source. All transparent. #CloudEngineering #DevSecOps 5/5

010

Matteo Depascale @matteodepascale.bsky.social · 27/02/2026

Key features: 🔹 Runs as DaemonSet on every node 🔹 Monitors: Kernel, Networking, Storage, Runtime, GPU 🔹 Auto-detects 5+ issue categories 🔹 Integrates with EKS auto-repair 🔹 Open source = full transparency Check it out on GitHub! #OpenSource #Kubernetes #AWS 4/5

120

Matteo Depascale @matteodepascale.bsky.social · 27/02/2026

Now with the open source EKS Node Monitoring Agent: ✅ Automated detection across 5 categories ✅ Kernel, network, storage, runtime, GPU monitoring ✅ NodeConditions for auto-repair integration ✅ Full visibility into detection logic #CloudComputing #DevOps 3/5

110

Matteo Depascale @matteodepascale.bsky.social · 27/02/2026

Before this agent: ❌ Manual node health checks ❌ SSH + dmesg guesswork ❌ Stale GPU errors going unnoticed for hours ❌ Manual monitoring across kernel, network, storage, runtime #DevOps #CloudNative 2/5

110

Matteo Depascale @matteodepascale.bsky.social · 27/02/2026

🚀 Big news for #Kubernetes users! Amazon EKS Node Monitoring Agent is now OPEN SOURCE! No more SSHing into nodes to debug pod failures. The EKS Node Monitoring Agent is now available on GitHub for everyone. #AWS #Kubernetes 1/5

100

Matteo Depascale @matteodepascale.bsky.social · 26/02/2026

One reframe: Stop: "how I built it" Start: "how I designed, owned, and drove impact" Been through an Amazon loop? What surprised you most? #AWS #SoftwareEngineering #TechCareers 6/6

000

Matteo Depascale @matteodepascale.bsky.social · 26/02/2026

Common mistakes: ❌ Tasks instead of ownership ❌ Vague stories, no data ❌ Dodging failure questions ❌ Wrong communication level 5/6

100

Matteo Depascale @matteodepascale.bsky.social · 26/02/2026

Prep that works: - 12-16 stories remixable across LPs - Map each to 3-5 LPs + different angles - STAR + Impact - Concrete metrics, tradeoffs, "what I'd change" 4/6

100

Matteo Depascale @matteodepascale.bsky.social · 26/02/2026

The Bar Raiser: - Outside the hiring team - Digs into shallow/polished stories - Ensures you meet the bar for your level You can't fake this one 3/6

100

Matteo Depascale @matteodepascale.bsky.social · 26/02/2026

Each interviewer covers 2-3 LPs. 10-12 LPs tested across the panel. Even technical rounds go deep on behavior: → Metrics → Alternatives → Consequences → Scope at your level, not just skills 2/6

100

Matteo Depascale @matteodepascale.bsky.social · 26/02/2026

Amazon loop interviews ≠ "just more interviews." 3-5 hours. Back-to-back. Every round tests Leadership Principles — even technical ones. Here's what actually happens 🧵 1/6

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

What's the most complex scheduling problem you've solved on AWS? #AWS #Serverless #CloudComputing 8/8

000

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

More tips: → Clean up one-time schedules after execution (they count against quotas) → Batch small tasks to maximize free tier → Templated targets for Lambda/SQS/SNS, universal for everything else 7/8

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

Pro tips: → Flexible time windows for non-critical tasks → DLQs on every schedule, not just critical ones → Monitor InvocationAttempts vs SuccessfulInvocationAttempts ratio 6/8

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

Built-in resilience: → At-least-once delivery guarantee → 185 retry attempts over 24h w/ exponential backoff → DLQ integration for anything that still fails No more custom retry logic from scratch 5/8

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

Universal target support: → 270+ AWS services, 6K+ API operations → Templated targets for Lambda, SQS, SNS, Step Functions → Universal targets for any supported AWS API 4/8

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

3 schedule types out of the box: → rate-based for regular intervals → cron-based with timezone + DST handling → one-time for precise single executions 3/8

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

The old way: ❌ Max 5 TPS ❌ 20 targets ❌ 300 rules per account ❌ Custom retry logic EventBridge Scheduler: ✅ Thousands of TPS ✅ 270+ services, 6K+ API ops ✅ 10M schedules per region ✅ 185 retries over 24h built-in 2/8

100

Matteo Depascale @matteodepascale.bsky.social · 25/02/2026

Stop struggling with AWS scheduling at scale Still using cron jobs and custom polling loops? There's a better way → Amazon EventBridge Scheduler Here's why you should switch 🧵 #AWS #Serverless 1/8

110

Matteo Depascale @matteodepascale.bsky.social · 24/02/2026

2026 default stack: → CloudFront + TLS 1.3 + HTTP/2/3 + HTTP→HTTPS redirect → ACM for auto cert management → ALB/NLB with TLS listeners → PrivateLink/VPC endpoints for private traffic No excuses left #cloud #security #softwareengineering 5/5

010

Matteo Depascale @matteodepascale.bsky.social · 24/02/2026

Still valid pushback: → Microsecond-latency trading systems → Constrained IoT/embedded devices → Legacy systems where retrofitting TLS has real cost Otherwise? Encrypt everything 4/5

100

Matteo Depascale @matteodepascale.bsky.social · 24/02/2026

Edge/CDN termination puts the handshake close to the user Crypto offloaded from your origin Actual CPU cost? ~2% at peak. Noise on modern hardware 3/5

100

Matteo Depascale @matteodepascale.bsky.social · 24/02/2026

The real overhead was never CPU It was handshake round-trips TLS 1.3 cuts those in half. 0-RTT resumption for repeat clients HTTP/2+3 means you rarely pay that cost anymore 2/5

100

Matteo Depascale @matteodepascale.bsky.social · 24/02/2026

"HTTPS everywhere is unnecessary overhead" is a 2010 argument In 2026, it's mostly a myth Here's why 🧵 1/5

100

Matteo Depascale @matteodepascale.bsky.social · 23/02/2026

How to enable nested virtualization on EC2: New instance → Advanced details → Nested virtualization → Enable Existing instance → Stop → Actions → Instance settings → Change CPU options → Enable That's it. #AWS #DevOps #CloudComputing 4/4

000

Matteo Depascale @matteodepascale.bsky.social · 23/02/2026

Before you enable it — know the gotchas: ⚠️ Windows: VSM auto-disabled ⚠️ No hibernate/resume support ⚠️ Windows capped at 192 vCPUs (no m8i.96xl) ⚠️ Latency-sensitive workloads → still use bare metal ⚠️ Security inside nested VMs = your responsibility 3/4

100

Matteo Depascale @matteodepascale.bsky.social · 23/02/2026

Use cases this unlocks on regular EC2 instances: → Docker Desktop / WSL2 → Android Studio emulators → QEMU-based dev → In-vehicle hardware simulation → Mobile app emulator testing No more paying for bare metal just to run an emulator 2/4

100

Matteo Depascale @matteodepascale.bsky.social · 23/02/2026

Nested virtualization on EC2 is now available on virtual instances, no bare metal required Nitro passes Intel VT-x extensions through 3 layers: L0: AWS/Nitro → L1: Your EC2 (KVM/Hyper-V) → L2: Your nested VMs Supported: C8i, M8i, R8i. Zero extra cost 🧵 #AWS #EC2 1/4

100

Matteo Depascale @matteodepascale.bsky.social · 20/02/2026

EventBridge pattern constraints you need to know: → 2,048 chars per pattern → 1,000 rule combinations ($or limit) → 30 wildcard rules per event bus → Numeric range: -5.0e9 to +5.0e9 Push logic into rules, keep Lambda for actual processing #AWS #EventBridge #Serverless 4/4

000

Matteo Depascale @matteodepascale.bsky.social · 20/02/2026

EventBridge filtering implementation tips: • source + detail-type as primary filters • Numeric operators for threshold monitoring • $or for cross-field logic (max 1,000 combos) • Test with EventBridge Sandbox • Monitor TriggeredRules metrics for anomalies 3/4

100

Matteo Depascale @matteodepascale.bsky.social · 20/02/2026

Common EventBridge filtering gotchas: → Confusing field-level AND with value-level OR (#1 source of bugs) → Case-sensitive matching is the default → Overusing wildcards when prefix/suffix operators work fine → One massive rule instead of multiple precise ones 2/4

100

Matteo Depascale @matteodepascale.bsky.social · 20/02/2026

You don't need Lambda to filter EventBridge events EventBridge has a dual boolean logic system most people underuse: → Implicit AND across fields → Implicit OR within arrays → Explicit $or for cross-field conditions No filtering code needed 1/4

100

Matteo Depascale @matteodepascale.bsky.social · 19/02/2026

Setting up AWS Backup for EKS? Don't skip these: → Enable EKS in AWS Backup Settings first → IAM role needs AWSBackupServiceRolePolicyForBackup → Add S3 permissions if backing up S3 buckets → Cluster auth mode must be API or API_AND_CONFIG_MAP #AWS #Kubernetes 4/4

000

Matteo Depascale @matteodepascale.bsky.social · 19/02/2026

AWS Backup for EKS — key capabilities: • Backs up cluster state (deployments, resources) AND persistent data (EBS, EFS, S3) • Restore full clusters, namespaces, or individual PVs • Immutable vaults for compliance • Single centralized policy across all AWS services 3/4

100

Matteo Depascale @matteodepascale.bsky.social · 19/02/2026

What EKS backup used to look like: ❌ Custom scripts per cluster ❌ Extra licensing for third-party tools ❌ Manual orchestration for state + PVs ❌ Provision a new cluster just to restore AWS Backup for EKS fixes all of this natively 2/4

100

Matteo Depascale @matteodepascale.bsky.social · 19/02/2026

AWS Backup now supports Amazon EKS natively — no agents, no scripts, no third-party tools. One policy. Full cluster state + persistent volumes (EBS, EFS, S3). Cross-region & cross-account copies. Still scripting your own EKS backups? 🤔 #AWS #Kubernetes 1/4

100

Matteo Depascale @matteodepascale.bsky.social · 18/02/2026

Note: For deepest reasoning tasks (complex refactoring, multi-agent coordination), Opus 4.6 still leads. Already testing Sonnet 4.6? What are you seeing? Drop your findings below 👇 #CloudComputing #SoftwareEngineering 6/6

010

Matteo Depascale @matteodepascale.bsky.social · 18/02/2026

🔹 Enterprise Ready - Spreadsheet & financial model creation - Compliance review with attention to detail - Data summarization where speed + accuracy matter Migration from 4.5? Minor prompt adjustments. Supports adaptive + extended thinking. Drop-in upgrade for most cases. 5/6

100

Matteo Depascale @matteodepascale.bsky.social · 18/02/2026

🔹 Agentic Workflows Works as lead agent OR subagent in multi-model pipelines. Better at multi-step orchestration, less prone to overengineering and "laziness" than 4.5. Precise workflow management out of the box. 4/6

100

Matteo Depascale @matteodepascale.bsky.social · 18/02/2026

🔹 Context Window 1M tokens in beta. Entire codebases, lengthy contracts, dozens of research papers in one request. Auto-summarizes older context as conversations grow—no more manual chunking. 3/6

100

Matteo Depascale @matteodepascale.bsky.social · 18/02/2026

🔹 Computer Use Best model Anthropic has shipped for browser automation. Navigate spreadsheets, multi-step forms, multiple tabs with near-human reliability. OSWorld benchmarks show steady gains over every prior Sonnet. 2/6

100

Matteo Depascale @matteodepascale.bsky.social · 18/02/2026

Claude Sonnet 4.6 is now on Amazon Bedrock 🚀 Frontier AI at $3/$15/M tokens—near-Opus performance at a price that actually scales. Here's what changed for builders 🧵 #AWS #AI 1/6

100

Matteo Depascale @matteodepascale.bsky.social · 17/02/2026

The #1 mindset shift: Events = business facts ("OrderConfirmed") NOT state changes ("row_updated") Rules match business concepts, not technical boundaries. Teams own their domain bus and evolve independently. #AWS #EventBridge #ServerlessArchitecture 5/5

000

Matteo Depascale @matteodepascale.bsky.social · 17/02/2026

Bus-first checklist: → One custom bus per bounded context → Events as immutable past-tense facts, never commands → DLQ on every single target → Schema Registry from day one → Reverse-DNS source naming (com[.]ecommerce[.]orders) 4/5

100

Matteo Depascale @matteodepascale.bsky.social · 17/02/2026

Anti-patterns killing your EventBridge setup: ❌ CRUD sourcing (Inserted/Updated/Deleted) ❌ Property sourcing (guest_first_name_changed avalanches) ❌ Clickbait events forcing sync queries back to producers ❌ Everything on the default bus 3/5

100

Matteo Depascale @matteodepascale.bsky.social · 17/02/2026

Source-first = tight coupling, rule sprawl, silent failures Bus-first = loose coupling, independent teams, comprehensive DLQs with 24hr retries The difference? Bus-first treats event buses as domain containers, not service wrappers. 2/5

100