Web Application Development

Software that carries the weight of your business

We build the product your operation actually runs on — typed end to end, tested where it counts, and documented so your next developer starts on day one, not month three.

From $28,000Typical timeline 8–16 weeksProducts that hold up under real load
Median uptime across shipped apps
99.98%Median uptime across shipped apps
Typical time to first production release
6–14 wksTypical time to first production release
P95 API response we design toward
<200msP95 API response we design toward
The problem

Most teams come to us at one of these moments.

None of these are unusual, and none of them mean anyone did a bad job. They are what happens when a business grows faster than the systems holding it up.

Our first job is to work out which of these is actually costing you money — and to tell you if the answer is none of them.

  • A prototype grew into the business and now nobody dares touch it.

  • Three internal tools, three logins, and a spreadsheet holding it together.

  • Every new feature breaks two old ones because nothing is tested.

  • Your last agency delivered code no one on your team can read.

What we do

Inside web application development.

Six areas we cover in depth. You do not have to take all of them — most engagements start with two or three.

SaaS platforms

Multi-tenant architecture, role-based permissions, billing and usage metering wired in from the first sprint — not bolted on later.

Data-heavy dashboards

Interfaces that stay fast at a million rows: server-side pagination, streamed queries, and caching that reflects how people actually read data.

APIs & integrations

REST and GraphQL services, webhook pipelines, and the unglamorous plumbing between your CRM, ERP and payment provider.

Payments & subscriptions

Stripe and Paddle flows including trials, proration, dunning and tax — the edge cases that quietly cost you revenue.

Progressive web apps

Installable, offline-tolerant experiences that behave like native software without the two-app-store tax.

Legacy modernisation

We replace the old system module by module, so your business keeps running while the foundation changes underneath it.

What you get

Everything listed here is in the contract.

No line items that turn out to be optional extras later. If it is on this list it is scoped, priced and delivered.

  • Technical architecture document
  • Typed codebase, end to end
  • Component & design system
  • Automated test suite
  • CI/CD pipeline
  • Staging + production environments
  • Monitoring & error tracking
  • Handover session + written docs
Tools & stack

We pick boring, well-supported tools on purpose. Everything below is something your next hire can already use.

  • TypeScript
  • Next.js
  • React
  • Node.js
  • PostgreSQL
  • Prisma
  • Redis
  • tRPC
  • AWS
  • Docker
  • Playwright
  • Stripe

You own every account, repository and licence we set up on your behalf.

The process

How a web applications engagement runs.

Fixed phases with a defined deliverable at the end of each. You approve one before the next begins.

  1. 01Week 1–2

    Technical discovery

    We map the domain, the data and the constraints — then write down the architecture before a line of product code exists.

  2. 02Week 2–3

    Foundations sprint

    Repo, environments, auth, design system and deployment pipeline. Everything after this ships to a real URL.

  3. 03Week 3–12

    Feature sprints

    Two-week cycles with a working demo at the end of each. You see progress in the product, not in a status deck.

  4. 04Final 2 weeks

    Harden & hand over

    Load testing, security review, docs, and a walkthrough with your engineers so ownership genuinely transfers.

Case study

What this looks like in practice.

Corvid Logistics — Freight & Supply Chain. 16 weeks of work, with the numbers to show for it.

Rebuilding a live scheduling engine without stopping the trucks

Corvid ran 2,000 daily dispatches through a PHP monolith written by a contractor who had long since left. Adding a customer took an engineer half a day. Peak-hour page loads exceeded eleven seconds, and dispatchers had started keeping a parallel spreadsheet — which meant the system of record was no longer the system of record. A rewrite was necessary, but a big-bang cutover was unthinkable: a single lost dispatch day costs more than the entire project budget.

Days of downtime during migration
0Days of downtime during migration
Peak dispatch board load time
11s → 0.9sPeak dispatch board load time
Reduction in support tickets
61%Reduction in support tickets
Faster onboarding of new customers
Faster onboarding of new customers
Read the full case study
References

Clients who bought exactly this.

Every quote below is from a client of this specific service. We will happily put you on a call with any of them before you sign anything.

We had eighteen months of technical debt and a board deadline. OnyxEra rebuilt the core scheduling engine while the old one kept running — not a single day of downtime, and our support tickets dropped by more than half.
61% fewer support tickets
Priya RaghunathanVP Engineering, Corvid Logistics
They asked harder questions in the first week than our previous vendor did in a year. The architecture document alone was worth the engagement — we still make decisions against it.
Daniel OkaforFounder & CTO, Sift Health
Our platform went from 40 to 4,000 concurrent users during a launch and nothing bent. That is entirely down to how they built it.
100× traffic, zero incidents
Marta LindqvistHead of Product, Northbeam Studio
FAQ

Web Applications, answered straight.

If something is not answered here, ask us directly — you will get a straight answer rather than a sales call.

Yes, and roughly a third of our work starts that way. We begin with a paid two-week audit: we read the code, run it, document what we find, and give you an honest recommendation — refactor, rebuild, or leave it alone. You keep that document whether or not you continue with us.

Web Applications

Ready to talk about web applications?

Bring the messy version of the problem. Thirty minutes on a call is usually enough for us to tell you what it would take and roughly what it would cost.

Typical reply time: under 4 business hours