Services

Everything needed to take software from idea to scale

Five practices, one team, one contract. Engage us for a six-week discovery, a full product build, or as the platform team you never got round to hiring.

Abstract visualisation of a global network
01 — Cloud & platform

Infrastructure that stops being a topic of conversation

Most teams do not need more servers. They need a platform where deploying is boring, rollbacks are instant, and the on-call rotation is quiet.

  • Multi-region AWS, GCP and Azure architecture
  • Kubernetes, Terraform and GitOps delivery pipelines
  • Cost engineering — clients average a 34% cloud bill reduction
  • Observability with tracing, SLOs and alerting that means something
  • Zero-downtime migrations off legacy hosting
Abstract representation of an artificial intelligence system
02 — Applied AI & data

AI features that survive contact with real users

Demos are easy. We build the unglamorous parts — evaluation harnesses, guardrails, fallbacks and cost controls — that decide whether a model makes it to production.

  • Retrieval-augmented search over your own documents and data
  • LLM workflow automation with human review where it matters
  • Forecasting, ranking and anomaly detection models
  • Offline and online evaluation, with regression suites per release
  • Data pipelines, warehousing and governance
Designer sketching interface wireframes
03 — Product & design

Interfaces people do not have to be trained on

Design at Basalt is not decoration applied at the end. Our designers sit in the same standup as the engineers and ship in the same sprint.

  • User research, journey mapping and problem framing
  • Interactive prototypes tested before code is written
  • Design systems delivered as real, documented components
  • WCAG 2.2 AA accessibility built in, not retrofitted
  • Front-end engineering in React, Next.js and Svelte
Also available

Two more ways we help

04 — Security & compliance

Threat modelling, penetration testing and the evidence trail auditors ask for. We have taken eleven clients through SOC 2 Type II without a single qualified finding.

  • SOC 2, ISO 27001, HIPAA and GDPR readiness
  • Application and infrastructure penetration testing
  • Secure SDLC, secrets management and access review

05 — Managed operations

Once it is live, someone has to run it. Our operations retainer gives you a named engineer, a 15-minute severity-one response, and a monthly reliability report.

  • 24/7 monitoring and incident response
  • Dependency patching and quarterly upgrade windows
  • Capacity planning ahead of your growth, not after it
Toolkit

The stack we reach for

Chosen for maturity and hiring pool, not for novelty. We will happily work in yours instead.

TypeScriptReactNext.jsGoPython RustPostgreSQLKubernetesTerraformKafka
Engagement models

Three ways to work with us

Indicative pricing. Every proposal is scoped properly before you are asked to sign anything.

Discovery sprint

For teams who need a costed plan before committing budget.

$18k/ fixed, 2 weeks
  • Architecture and codebase review
  • Stakeholder and user interviews
  • Clickable prototype of the core flow
  • Costed delivery roadmap
  • Yours to keep, no obligation
Book a sprint

Operations retainer

For live systems that need a team on call without a headcount req.

$9k/ month
  • 24/7 monitoring and incident response
  • 15-minute severity-one response time
  • Named engineer who knows your stack
  • Monthly reliability and cost report
  • Rolling 30-day term
Talk to us
FAQ

Questions we get asked first

Something not covered? Email hello@basalt.dev and a real engineer will answer.

Discovery sprints usually begin within two weeks. Full build pods depend on availability — typically four to six weeks out. If we cannot staff you properly, we will say so rather than start with the wrong people.

You do, entirely and from day one. Everything is committed to your repositories under your accounts. There are no licence fees, no proprietary frameworks and nothing you would need our permission to change later.

That is most of what we do. We join your standups, your board and your repositories. Roughly half our engagements are explicitly structured so your engineers take full ownership at the end.

Scope always changes. Build pods are billed monthly rather than per feature, so re-prioritising costs nothing but a conversation. We will flag the trade-off in the sprint it happens, not in a change-request letter three weeks later.

Yes to both. We are SOC 2 Type II certified, carry professional indemnity cover, and our counsel can usually turn a standard MSA around in under a week.

Not sure which one you need?

Most conversations start with a thirty-minute call and no slide deck. Describe the problem and we will tell you what it would actually take.