Skip to main content
Security

We get access to your crown jewels. Here's how we treat them.

Our work touches ERPs, financial data, and customer records. These are the practices we follow on every engagement.

Practices

Least-privilege access

Scoped accounts per engineer, removed when the engagement ends.

Encryption

TLS in transit and encrypted storage for data we handle.

Secure development

Code review, dependency scanning, and secrets kept out of repositories.

Data minimization

Anonymized copies for development and testing whenever possible.

NDAs and agreements

Confidentiality from day one, and we'll review your security addendum.

Incident response

A defined process to contain, notify, and review security incidents.

Need a security questionnaire answered?

Send it over. We'll answer honestly, including where something is your responsibility or a third party's.

Security as a small team

Being a startup doesn't change what's at stake: we work with ERPs, financial records, and customer data, and a mistake would hurt you far more than us. So we keep our practices simple, consistent, and applied to every engagement, regardless of its size. We don't claim certifications we haven't earned; we explain exactly what we do and let you judge.

A small team is also an advantage here. Fewer people need access, every engineer is known personally, and access is easier to review and remove.

How access is handled

SituationOur practice
Starting a projectNamed accounts per engineer, with the minimum permissions needed
Production systemsAccess only when required, logged, and preferably through your tools
Test and developmentAnonymized or synthetic data wherever possible
CredentialsStored in a password or secrets manager, never in code or chat
End of engagementAll access removed and confirmed in writing

If something goes wrong

If we suspect a security incident affecting your systems or data, we contain it first, tell you promptly with what we know, and work with you on recovery. Afterwards, we write a short review covering the cause and the changes that prevent it from happening again.

What you can ask us for

  • A completed security questionnaire
  • A mutual NDA before discovery
  • Review of your security or data processing addendum
  • A list of everyone with access to your systems
  • Confirmation of access removal at the end of the project

Secure development practices

Every change goes through code review before it reaches your systems. Dependencies are scanned for known vulnerabilities, secrets are kept out of repositories with automated checks, and production deployments run through pipelines rather than manual changes on servers. These habits apply equally to a small integration and a full product build.

For systems that handle sensitive data, we also review access rules, logging, and data retention with you during design, so security decisions are made deliberately rather than inherited from defaults.

Who is responsible for what

AreaUsually usUsually you
Code we writeSecure design, review, and fixesApproving releases
Your infrastructureConfiguration we're asked to manageAccounts, billing, and policies
Third-party servicesIntegrating them securelyContracts and vendor reviews
Your usersRole and permission designGranting and removing access
IncidentsContainment and technical recoveryNotifying customers or regulators

Third-party services

Projects often rely on hosting providers, payment gateways, email services, or AI model providers. We help you choose services whose security and data terms fit your requirements, configure them with least privilege, and document where your data goes, so nothing is a surprise later.

Access hygiene we follow

  • Personal accounts for every engineer, never shared logins
  • Multi-factor authentication on every client system we can enable it on
  • Least-privilege access, granted per project
  • Access removed promptly when work ends
  • Secrets stored in a password manager or secret store, never in chat or email

Least privilege, in practice

We ask for the minimum access a task needs, for the shortest time it's needed. Production access is the exception, not the default: most work happens in staging environments with realistic but anonymized data where possible.

Each engineer uses their own named account, never a shared login, so every action is attributable. When an engagement ends, or a person moves off a project, their access is removed and you get confirmation.

Secrets and credentials

Passwords, API keys, and certificates never live in source code or chat messages. They're stored in a secrets manager or the platform's encrypted configuration, rotated when people change roles, and scoped so that a leaked key can do as little as possible.

Integrations get their own service accounts with only the permissions they need. If one is compromised, it can't be used to reach unrelated data.

Backups you have actually restored

A backup is only useful if it can be restored, quickly, by someone who isn't the person who set it up. We automate backups, store copies separately from the systems they protect, and schedule restore tests so the recovery steps are proven rather than assumed.

For critical systems we agree recovery targets with you in advance: how much data you could afford to lose, and how long the system can be down. Those targets then drive the backup frequency and the standby setup.

Ask us anything

If your security team has a questionnaire, send it over. We'll answer honestly, including where something isn't in place yet.

Frequently asked questions

Can you work under our security policies?
Yes. We'll follow your access, device, and data-handling policies, and we're happy to complete security questionnaires before an engagement starts.
Do you use client data to train AI models?
No. Client data is used only for your project. When AI services are involved, we choose configurations and providers that don't train on your data and document that choice.