I am Igal Schneider, an AWS Certified Solutions Architect in Toronto. For 20+ years I have designed, built and operated production cloud and Kubernetes estates for teams where downtime and runaway cost are not options. I own the decisions, not just the advice.
Cloud architecture, DevOps automation and FinOps, under one owner who is accountable for cost, reliability and security from end to end.
Design and operate scalable, highly available estates on AWS. From greenfield to live migrations under traffic, with the boring parts done right.
Run production EKS the way it should be run: zero-downtime rolling upgrades, PodDisruptionBudgets, and control-plane and node migrations under live traffic.
Centralized, automated delivery across Jenkins, Bitbucket Pipelines and Bamboo. Infrastructure as code, slimmer images, faster and safer deploys.
Find the leak before it becomes a budget. Forensic analysis of access logs and usage, then re-architecture that cuts spend without touching the product.
Operate the full delivery stack for AVOD and connected-TV at consumer scale: S3 + CloudFront, multi-CDN strategy, and HLS playback that just works.
Own IAM, TLS and certificate lifecycles, WAF, and security-risk monitoring as code, so the safe path is also the default path.
Not demos. Real software running in production clusters and customer environments, built with LLM-driven workflows as a genuine force multiplier.
An AI-driven audit agent that automates cluster hygiene and surfaces misconfigurations. Database-backed service with migrations and an automated audit pipeline that reduces operational toil.
A Go-based in-cluster Kubernetes audit tool with its own backend, web UI and deployment. Runs inside production clusters to surface posture and configuration issues in real time.
A Zabbix and Datadog-synthetics replacement: 11+ check types including declarative browser-synthetic checks and HLS video-playback monitoring, customer-facing health dashboards, TOTP MFA and Fargate deployment.
LLM-driven workflows across architecture reasoning, code generation and incident diagnosis. Prompt engineering treated as seriously as any other engineering skill.
A few signature results from operating real systems under real constraints, not the ideal ones.
Executed with S3 Batch Operations alongside cross-account 12 to 22 TB video-asset transfers on high-bandwidth compute, with no service interruption.
Forensically analyzed S3 access logs in Athena and re-architected the tiering strategy. An $84K+/year FinOps win, delivered without touching the product.
Migrated two production clusters off self-hosted kops onto EKS, then owned control-plane and node-OS upgrades under live traffic with zero-downtime rollouts.
When new TLS requirements broke a legacy embedded-device fleet, engineered a proxy service to restore consumer delivery in production while a long-term fix was built.
Built pre-aggregated daily tables with DISTKEY/SORTKEY and materialized views, and tracked down a runaway query holding a cluster at sustained 100% CPU.
Decomposed a monolithic application into config, programming and UI services, so teams could ship independently without stepping on each other.
A structured path from first look to long-term ownership, the way a technical co-founder would run it.
Understand the real constraints, map the current estate, and find the costliest and riskiest gaps first.
Design against your actual environment with AWS best practices, then sequence the work to minimize blast radius.
Execute with infrastructure as code and zero-downtime rollouts. Live systems stay live while they modernize underneath.
Wire up the pipelines, guardrails and observability so the safe path is the fast path.
Right-size, re-tier and remove waste. Performance and cost get measured, not guessed.
Stay long enough to make sure it keeps working, modernizing it as the business grows.
Streaming, connected-TV and technology companies across North America.
Before the consultancy, a career spent inside and alongside some of the largest technology and enterprise organizations in the world.
Worked at
Delivered for global enterprises
Enterprise deployments, network-traffic analysis and on-site delivery across North America, Europe and Asia, under formal SLA and ISO process discipline.
I solve hard cloud problems, and I stick around long enough to make sure they stay solved. Over 20+ years I have built and run production systems for organizations where downtime and runaway cost are not options. My work sits at the intersection of cloud architecture, DevOps automation and FinOps, because building something that scales is only half the job. The other half is making sure it does not drain your budget while doing it.
My defining trait is tenure with depth. I have been embedded inside a single company since 2012, operating its live AVOD streaming platform (in production since 2016) and running its full AWS and Kubernetes estate through every major platform shift. That is the kind of long, single-environment ownership that teaches you exactly where the bodies are buried, and how to keep a system alive while modernizing it underneath live traffic.
Through IT Delivered I work with technology companies and startups as their embedded cloud expert, the person who owns the infrastructure decisions, not just advises on them. I am in the weeds: tuning Kubernetes clusters, debugging Redshift queries, and finding the cost leak before it becomes a six-figure problem.
I have moved decisively into AI-augmented engineering, shipping real production software through AI-assisted development rather than treating it as a novelty. CloudScope, ik8sa and Watchpost were all designed and built this way. I use LLM-driven workflows as a force multiplier across architecture reasoning, code generation and incident diagnosis.
Small, fast, no-signup utilities I built and keep online. Handy when you are knee-deep in a deploy.
In the best possible way. Reliable, cost-efficient, and quietly doing its job. Reach out and let's talk.
