Eorbitt — home
Book a technical call

Capability

Software engineering

You have a process running on spreadsheets, a vendor tool that does four fifths of the job, or an application written by someone who left. It works until it does not, and nobody wants to touch it.

At a glance

Phases
5
First phase
1 to 2 weeks
Deliverables
8
Stack layers
5
Book a technical call

What we deliver

The work itself

  • Map the existing process end to end, including the exceptions people handle by hand
  • Design a data model that matches how the business actually works, not how the old tool forced it
  • Build the application in TypeScript against a typed schema, with migrations under version control
  • Cover the paths that carry money or compliance risk with integration tests, not unit-test theatre
  • Ship behind a deployment pipeline that any engineer on your side can read and run
  • Instrument the application so failures surface as alerts rather than as support tickets
  • Write the runbook and the architecture decision records alongside the code
  • Migrate live data with a rehearsed cutover and a tested path back

How we work

What the engagement looks like

Phases, durations and what each one hands over are listed below. The stack is published in full — nothing about how this is built is a surprise once work starts.

Stack

What we build it with

  • Interface
    TypeScript · React · Next.js (App Router) · Tailwind CSS
  • Runtime
    Node.js · Hono · REST and typed RPC · Zod schema validation
  • Data
    PostgreSQL · Drizzle ORM · Redis · Object storage (S3 or R2)
  • Platform
    Vercel · Cloudflare Workers · Azure App Service · Docker
  • Operations
    GitHub Actions · Playwright · Vitest · OpenTelemetry · Sentry

Engagement shape

Phases and timeline

  • Discovery
    1 to 2 weeks
    Process map, data model, risk register, and a build estimate with the assumptions written down
  • Architecture
    1 week
    Architecture decision records, schema, environment plan, and the test strategy
  • Build
    6 to 16 weeks in two-week sprints
    Working software in a staging environment at the end of every sprint
  • Cutover
    1 to 2 weeks
    Rehearsed migration, parallel run, and a rollback that has been executed at least once in staging
  • Handover
    1 week
    Runbook, architecture notes, credential transfer, and a working session with your engineers

What we need from you

  • One decision-maker who can settle scope questions inside two business days
  • A subject-matter expert for four hours a week during discovery and build
  • Read access to the system being replaced, including its database if one exists
  • A named owner on your side for the credentials we will hand back

After go-live

Who holds it at 3am

  • Alerts route to an on-call rota, with the first response inside the window set in the contract
  • Dependency and security updates are applied on a fortnightly cycle, not when something breaks
  • A monthly note covers error rates, slow queries, and the work we recommend next
  • The pipeline, the runbook and the credentials are yours; nothing is held hostage as leverage