DevOps as a Service, built for startups
Ship faster. Spend less. Sleep better.
Your developers are losing time and runway to cloud setup, security gaps and a budget nobody is watching, instead of shipping features. We set up the guardrails in your own AWS account, then run them for you on a monthly subscription.
We build it
Live in weeks
We run it
Every week after
Subscription
Not a salary
Read only access · No commitment · You keep the report
Straight to Humayun, not a sales team.
Sound familiar?
The wall almost every early stage startup hits.
Best fit
Running on AWS, one to three environments, no dedicated DevOps hire yet, and infrastructure that has outgrown whoever set it up.
Too early if
You are pre product or pre revenue with one environment and no customers. There is not enough there yet to be worth guarding.
Too late if
You already have a platform team. You have solved this a different way, and we would only get in their way.
Know where you stand before you spend anything
Two free checks, each answering one question about your AWS. Pick whichever is hurting. Read only access, a report in five business days, and a 30 minute walkthrough. You keep the report either way.
Free security check
Would your AWS pass an enterprise security review?
A straight answer on whether you would pass. Your account scanned, then every finding checked by hand rather than handed to you as raw tool output.
Your top risks, in the order they would cost you a deal. A short report you keep, ranked by what to fix first.
Free cost check
How much of your AWS bill is waste?
A number you can act on. What you are spending on resources nobody is using, oversized instances and storage nobody has looked at since it was created.
Each saving with its effort and its risk. So you can tell the easy wins from the ones that need a maintenance window.
How it works, either way
Book a 30 minute call
We put your walkthrough in the diary on that same call, so you leave with a date.
Give read only access to your AWS account
I'll guide you through it. Nothing in your account is changed.
I check the account by hand, within 5 business days
A scan first, then every finding checked by a person, not handed to you as raw output.
We walk through it together, 30 minutes
You get the report on that call, and you keep it whatever you decide next.
I run a small number of these each month, because each one is a person reading your account rather than a tool emailing you a PDF. If you are not a fit I will say so on the first call, and you will have lost 30 minutes rather than a fortnight.
Fixed scope: one AWS account, and one question. The security check looks at security, the cost check looks at cost. Covering every account you run, across all six pillars, is the paid assessment below.
Nothing, unless you want it to. If you do want the findings fixed, that is quoted as a fixed price before any work starts: for cost, priced against the savings we found, so it pays for itself; for security, priced on the specific bugs you agree are worth fixing. You see the number first and you decide.
The alternative is finding out the hard way
Wait and find out
The questionnaire arrives, or the bill does. You fix it with a customer waiting on the answer, and the deal slips a quarter.
Run a tool yourself
Hundreds of findings, no ranking, and no way to tell which ones a reviewer would actually stop on. Your developers lose a week triaging it.
A free check
A ranked list of what actually matters, checked by a person, in five business days. Free, read only, and yours to keep.
A 30 minute call to set it up. Nothing to prepare.
Want the detail first? See what each step costs or read the common questions.
Cover the seat until your hire starts
Hiring a senior DevOps or platform engineer can take months, and none of it waits for them. A developer is already covering AWS between features, the bill keeps climbing, security questionnaires sit unanswered, and every deploy is a little slower than the last.
That is just as true whether the role is open now or still a line in next quarter's plan. The gap is already costing you.
I keep your foundation healthy until your hire starts, then hand over everything they need to take it from day one.
What I cover
What it is not
I do not replace the hire, and I do not cover on call. The goal is a clean handover.
How it works
Designed to end when your hire starts.
Everything runs in your own AWS account, so there is no contract to unwind and nothing of mine left behind to depend on.
The one page plan for covering a DevOps seat while you hire
What to protect first, what can safely wait, and how to get your new engineer productive in their first week. Useful even if you have not opened the role yet.
Tell me where to send it. No call required.
What is DevOps as a Service?
Traditional DevOps is something you build. You hire engineers, pick a toolchain, and spend months assembling pipelines, infrastructure code, monitoring and security controls yourself. DevOps as a Service is something you consume: an external team brings the toolchain and the expertise already built, sets it up in your cloud account, and runs it for you on a subscription.
You do not have full time work for a DevOps engineer.
You have full time risk.
Those are different problems, and hiring only solves the first. The work is real and it never stops. The bill drifts, dependencies age, access spreads, the pipeline breaks on a Friday afternoon. But it adds up to a few days a month, not a job. Every other function shaped like that got shared a long time ago, which is why nobody hires a full time accountant at this stage. Infrastructure is close to the last one still being solved by recruitment.
Why hiring goes wrong at this stage
Why a consultancy goes wrong too
A project has an end date. Someone arrives, builds, hands over a document and leaves. Six months later it has drifted, an AWS service has changed underneath it, and nobody still at the company knows how it fits together.
You did not buy a platform. You bought a snapshot of something that needed maintaining, and the maintaining is the part that was never included.
Four ways to solve it
Only one of them matches the size of the problem.
Hire and build it
A senior engineer, plus the months they need to build the platform.
Absorb it into the team
What most startups actually do, until it quietly stops working.
Buy it as a project
Built, handed over, and then nobody owns it.
Consume it as a service
Built once, then owned and maintained continuously.
It can be affordable because the work is nearly the same everywhere. The landing zone, the pipelines, the access model, the alerting, the backup policy: the shape is standard and the differences are configuration.
Not cheaper people. The same work, done once instead of forty times.
Why startups choose the model
Early stage teams have the same cloud problems as large ones, on a fraction of the budget and none of the headcount. This is the shape that fits.
Senior expertise, no hiring
Work with engineers you could not realistically recruit or afford at this stage, from the first week, with no recruitment cycle to survive.
Live in weeks
The pipelines, infrastructure modules and monitoring already exist. We adapt proven pieces to your stack instead of building a platform from scratch.
Subscription, not salary
A predictable monthly fee instead of a fixed cost on your payroll. Scale it down in a quiet quarter, scale it up before a launch.
Your developers stay on product
Runway goes into the thing customers pay for. Your team stops losing days to pipeline debugging, IAM policies and surprise bills.
Scales as you grow
Modular by design. Add environments, regions, compliance controls and on call coverage when you need them, not before you can use them.
Best practice stays current
Patching, AWS service changes and new security guidance are handled continuously, so the setup does not quietly rot the way an untended one does.
The guardrails we set up
Five guardrails, delivered as infrastructure as code in your own AWS account, so the safe path is also the easy one for your developers. A typical startup baseline is running in two to three weeks.
Landing zone and IaC
Accounts, networking and environments laid out properly, defined as code from the start
CI/CD pipelines
Push to main, ship to production, with test gates on the way and a way back if it goes wrong
Security and compliance
The controls an enterprise buyer will ask about, in place before they ask
Monitoring and logs
Metrics, log aggregation and alerts, so problems reach you before they reach customers
Backup and restore
Automated backups, and a restore somebody has actually tested
The part most providers skip
Guardrails handed over and forgotten are worth very little six months later. This is the ongoing work that keeps them true, included in the monthly fee.
Cost stays under control
We watch the bill so it never becomes a surprise at the end of the month
Shipping stays easy
Pipelines that keep working, and someone to call when a deploy will not go out
Access stays tight
Who can reach what, reviewed on a schedule instead of after an incident
Nothing rots quietly
Upgrades handled while they are routine, not once they are an emergency
Incidents get handled
Alerts reach a team that knows the system, not whoever is nearest a laptop
A team your developers can ask
The cloud questions that used to cost your team a day now cost them a message
Where the line is
Written down before you sign, so neither of us has to argue about it later.
Included in the monthly fee
Quoted separately
Not refusals. Just work that gets its own scope and its own number, rather than quietly eating the retainer.
If it's the foundation, it's included. If it's a new thing built on top, it's a project.
Built on standard, portable tooling
No proprietary layer of ours to unwind. If you take it in house later, your engineers already know everything here.
The full picture, not just security
A free check looks at one account and one question. The assessment looks at every account and environment you run, across all six AWS Well Architected pillars: where your cloud is leaking money, risking downtime or about to break, and in what order to fix it.
That is the price for a single account setup. Larger estates cost more, and a free check is what sets your number: you see one fixed figure, with a delivery date, before you decide anything. Credited in full against onboarding if you go ahead within 90 days.
Nothing owed if you stop here.
A free check
One account, one question, 5 business days
AWS assessment
Every account, all six pillars
Take a free check first. If it finds nothing that worries you, you have lost nothing and you do not need the assessment.
A free check
Security or cost, on one account. A short report you keep.
We review all six pillars
Cost, reliability, operations and architecture, not just security.
Findings walkthrough
We talk you through what we found, together on a call.
Roadmap you own
A prioritized plan. Run it yourself, or have us implement.
Read only access · No commitment · You keep the report
What you'll walk away with
No scores, no jargon, no 200 page PDF to decode. You get a plain list of what we found, a call to walk through it together, and a simple plan you own.
Every finding, not just the top ones
Cost, security, reliability and operations, explained in plain language. The free check gives you the headline risks; this is the whole list, with what each one costs you.
A walkthrough of the findings
We get on a call and talk you through everything together, so you understand what matters and why. Nothing is left as just a document.
A simple roadmap you own
What to fix first, in priority order, with the effort each one takes. This is the part the free check does not include. Your team can run it, or hand it back to us to implement.
Measured against the AWS Well Architected pillars
AWS defines a well run cloud through six pillars. Your assessment reviews your environment against each one, so you know exactly where you stand and what to fix first.
Operational Excellence
Run, monitor, and continuously improve your systems and day to day processes.
Security
Protect your data and systems with strong identity, encryption, and guardrails.
Reliability
Recover from failures and meet demand without downtime or data loss.
Performance Efficiency
Use the right resources efficiently as your needs and technology evolve.
Cost Optimization
Run at the lowest sensible cost while still meeting your requirements.
Sustainability
Reduce the energy use and environmental footprint of your workloads.
How you buy it
One path, priced step by step, and you can stop at the end of any of them. Every engagement opens with an assessment, because we will not quote a monthly fee for an account we have never seen.
A free check
Security or cost, whichever is hurting. A read only look at one AWS account, a short report of your top findings within five business days, and a 30 minute walkthrough. One account and one question: everything wider is the assessment.
You do not have to buy the assessment to act on a free check. Fixing what it found is quoted as a fixed price before any work starts: cost work priced against the savings found, so it pays for itself, and security work priced on the specific bugs you agree are worth fixing.
AWS assessment
A fixed price review of every account and environment you run, against all six Well Architected pillars: what it costs, what is exposed, what breaks under load and what to fix first. Findings in plain language, a call to walk through them, and a prioritized roadmap. Wider and deeper than a free check.
Take the roadmap and run it with your own team. Nothing is owed, and we would rather you did that than bought something you do not need.
Build
We put the guardrails in, scoped from what the assessment found. Fixed price, quoted as a number rather than an hourly rate, typically two to three weeks.
We document it, train your team and step back. No subscription. You can move onto managed later, because the platform is identical either way.
We run it
A foundation that stays true: watched continuously, and kept current by someone who knows how it was built. The monitoring runs on its own; the hours are for the person acting on what it finds.
Machinery we set up once. It runs whether or not anyone looks at it, so it costs you nothing per month.
Anything a person actually does. This is the honest half: someone has to act on what the machinery reports.
The standard plan, and what most startups need
More hours, more accounts, out of hours cover, compliance work
Hours cover hands on engineering work and do not roll over. In a normal month this work does not come close to the cap, which is why the cap exists rather than an hourly rate. If you go past it we quote the extra before starting, and if it happens two months running we will say so and scope you a custom plan rather than quietly bill you.
Month to month, 30 days notice, no minimum term and no exit fee. Move between tiers the same way. We would rather earn the next month than trap you in eleven of them.
A shared Slack channel with the engineer who built your platform. No project manager in between, no ticket queue, no first line reading from a script.
Out of hours cover can be arranged separately.
Wherever you stop, the platform is built as code in your own AWS account and documented for your team. If you outgrow us and hire in house, your new engineer inherits a clean setup instead of a mystery.
What this has looked like before
Three engagements where the platform layer was missing. Same shape of problem a startup hits, at a larger size.
Environment setup from days to under an hour
A global automotive manufacturer
Every environment was assembled by hand from a wiki page. Standing one up took days of an engineer’s time, and no two came out quite the same, so a bug that showed up in one could not reliably be reproduced in another.
A multi account AWS foundation defined in Terraform and CDK. Account structure, networking, IAM baselines and deployment pipelines all provisioned from code, so creating an environment became a single pipeline run instead of a checklist.
A new environment now takes under an hour and comes out identical every time. Onboarding time for new engineers dropped by around 40%.
45% faster page loads for a streaming catalogue
A Nordic streaming service
A large media catalogue served straight from the origin, with media processing sitting in the request path. Pages were slow for users far from the origin, and release day traffic made it worse.
Event driven streaming workflows that moved media processing off the request path and into asynchronous pipelines, a cached delivery path in front of the catalogue, and monitoring that surfaced regressions before users felt them.
Page loads 45% faster, and the processing work no longer competes with serving the catalogue.
Releases 40% faster, and anyone on the team can ship one
A Nordic energy tech company
Releases went out by hand. Shipping was slow and risky, and each release needed the one person who knew the sequence, which made holidays and handovers a problem of their own.
Standardized CI/CD: automated build, test and deploy pipelines with safe rollbacks, applied the same way across services, so the route to production was identical whichever service you were shipping.
Releases around 40% faster, and any engineer on the team can run one without supervision.
Who you are working with
Humayun Zia, founder and cloud architect, Stockholm. 14 years in software, DevOps and cloud, and an AWS Certified Solutions Architect.
Since 2020 I have run my own consultancy, brought in where the platform layer was missing: for a global automotive manufacturer, a Nordic streaming service, an energy tech company and Sweden’s largest classifieds marketplace.
Most recently, as Senior DevOps Engineer at an international EdTech marketplace, I designed the AWS account structure for a new platform, enforced security guardrails across every account, put bot protection in front of production and brought cost under control with reserved capacity.
Danzo is that same work, packaged for startups. You work with me directly: I assess your setup, build the foundation and keep it running.
Humayun Zia
Founder & Cloud Architect
Stockholm, Sweden
Connect on LinkedInTerraform · CDK · CloudFormation
AWS Registered Partner
Stockholm · US business hours
About Danzo
Danzo is a Stockholm based AWS consultancy and an AWS Registered Partner. We build and run cloud infrastructure for startups: a multi account AWS foundation, CI/CD, security guardrails and cost control, set up in your own AWS account and then operated for you on a monthly subscription.
We take on a small number of clients at a time, on purpose. It is the only way one senior engineer can be genuinely accountable for your production account, and it is the part of this that a larger firm cannot sell you.
Why Danzo works this way
Working with funded startups, I kept running into the same arithmetic. The runway is fixed, and a full time DevOps hire spends a serious share of it on a role that does not have a full week of work in it. The infrastructure problem is real and it never goes away, but it arrives in bursts, while a salary is a fixed cost against demand that is not.
The other thing I noticed is that the problem was rarely bespoke. Account structure, pipelines, IAM and secrets, monitoring, budgets, backups: almost every company I worked with needed the same foundation, and almost every one of them was paying to have it invented from scratch.
So Danzo does not take the traditional route. Not a project that ends in a handover document, and not a hire you have to recruit, pay and eventually replace, but a standard set of guardrails built in your own AWS account and then run for you month to month. Standardizing the work is what makes it affordable. Adapting the model rather than defaulting to the old one is what makes it fit a company with runway to protect.
Straight to the engineer · Read only access · No commitment
Before you book
How the model works, what it costs, and what you own at the end
Still have questions?
Book the free security check, or just send a message and I will reply within one business day.