DelivOpsBook a call

Shlomo Sfez · in your Slack · Israel

Your team moves faster when someone owns the infrastructure.

Senior DevOps ownership for your AWS, end to end. Built right, written down and kept clean, so your team stops working around its infrastructure and starts shipping on it.

Book a 30-minute call A few companies at a time
Hillel Saal
“Fifteen minutes later, my code is running in production.”
Hillel Saal · Backend Software Engineer, Flycomm

#infra

Backend lead

prod deploy failed on the migration step and now half the pods are crash-looping. we are stuck

Shlomo Sfez

ShlomoDelivOps

Looking. The migration ran before the new pods were ready — I am reordering the pipeline so it waits, and re-running.

Shlomo Sfez

ShlomoDelivOps

Green. In production 09:55, all pods healthy. Added a check so a migration can never run ahead of its release again — PR up for you to read.

  • 🙏 3
  • 🚀 1

A Tuesday, illustrative. The other party is a role, never a named client.

Runs on
  • aws
  • GCP
  • Azure
  • Terraform
  • Pulumi
  • Kubernetes
  • GitHub Actions
  • ArgoCD
  • Grafana
Running production for

Why it's faster

Your engineers already know DevOps. What they don't have is someone who owns it.

  1. Full ownership, end to end

    I own the area: design, build, run, and answer when it breaks. Not someone you hand tickets to.

  2. Right from the start

    On a new company, the foundations go in correctly the first time. On an existing one, I bring order to what is there. Either way, nothing gets built twice.

  3. Nothing lives in one head

    Everything is in code and in writing. The infrastructure is clean, and it is yours.

  4. AI for speed, experience for judgment

    AI makes the work fast. Knowing what is right, and in what order, makes it correct.

The work

What lands in your repo and your Slack.

Five kinds of work. Which comes first depends on what I find in your account; each one ends as an object you can open.

  1. Terraform

    The first pull request

    Your infrastructure, in code you own

    • An infrastructure repo in your GitHub, owned by you
    • A written account review before anything changes
    • CI that plans on every PR
    How this goes
    acme / infrastructure · pull request #12
    open

    import: bring the production VPC and RDS under Terraform

    • terraform
    • import
    • no-op
    module.network.aws_vpc.prod# imported, not recreated
    module.network.aws_subnet.private[0..2]
    module.db.aws_db_instance.main# imported
    module.db.aws_db_parameter_group.main
    Plan: 0 to add, 0 to change, 0 to destroy. 6 resources imported.
    • terraform plan41s
    • tflint8s
    • checkov / policy19s
    Shlomo Sfez

    Shlomo commented

    Zero changes to production. It is now code in your repo, and the next PR can change it safely.

  2. Pipelines

    A pipeline run

    CI/CD for ECS, EKS or serverless that takes a merged pull request to production without a human in the path, with tests and rollback already wired in.

    See the pipeline run on How I work

  3. Monitoring

    An alert, and the morning after

    Alerts that page on customers, not CPU

    • Alerts on what customers feel, not CPU
    • A runbook linked from every alert
    • One dashboard per service, same shape
    How this goes

    #alerts

    Alerts

    🔴 checkout · p95 latency above 800ms for 5m · 3.1% of requests affected · runbook: /runbooks/checkout-latency

    Alerts

    🟢 resolved · on-call followed the runbook: reconciliation job paused, checkout recovered in 4m

    Shlomo Sfez

    ShlomoDelivOps

    Saw the night. The job now has its own node pool so it cannot evict checkout again — PR up, no downtime to apply.

    • 🙏 2

  4. Security

    The findings report

    Least-privilege IAM, encryption everywhere, audit logging with retention and a WAF in front — the baseline that makes the questionnaire boring and the audit an evidence exercise rather than a rebuild.

    See the findings report on How I work

  5. Cost

    The monthly cost message

    A bill somebody owns

    • A tagged account with a per-service breakdown
    • A monthly report written for founders
    • Budgets and anomaly alerts before the invoice
    How this goes

    #infra

    Shlomo Sfez

    ShlomoDelivOps

    March: $18,420, down 9% from February. By service: EKS $7,900 · RDS $4,100 · data transfer $2,300 · S3 $1,600 · other $2,520. Next: the cross-AZ transfer on the ingest workers, worth about $900 a month. Commitments still on hold until that settles.

    Founder

    first time I have understood our bill. go ahead.

Illustrative, and typical. Real pull requests, alerts and reports stay in clients' workspaces. Inference workloads included: GPU nodes on EKS, SageMaker, Bedrock. Every platform and tool →

In their words

What it’s like
working with me.

15 people, 7 companies.
Founders, architects and the engineers who share a Slack channel with me.

Now we deploy dozens of times a day with zero downtime.

Nir Cohen

CTO and Co-Founder at Helfy

When Shlomo says he’s on something, you can relax: it will happen properly.

Translated from Hebrew

Yinon Eliyahu

Development Lead at Flycomm

A unique ability to turn complex problems into simple, elegant, and creative solutions.

Rotem Elharar

Chief Architect at Helfy

With him on board, I didn’t feel a need for another DevOps.

Ziv Kaspersky

VP Engineering at Vinesight

Every recommendation, in their own words

Pricing

Three tiers, set by how much I carry.

Monthly, 30 days notice. Priced on a call, by your scope. Every tier covers the same services. Tiers differ by infrastructure scope, planned-work capacity and review cadence.

What separates the tiers, and the questions people ask →

Thirty minutes, no deck.

Tell me what's breaking and I'll tell you honestly whether you need me, someone else, or nobody yet.

Book a 30-minute call Email me instead

A few companies at a time