Growth Engineer

About us

Velocity Growth is a global growth consultancy that combines growth strategy, market research, performance marketing, SEO, analytics, data science, pricing strategy, CRO, product-led growth, UI/UX design, and CRM systems into a single integrated engagement.

Backed by a team with growth leadership experience from Uber, Uber Eats, Accenture, Botim, Noon, and SWVL, the Velocity Growth team operates across 30+ markets spanning the GCC, Europe, UK, US, Asia Pacific and Australia.

Unlike traditional agencies that sell isolated services, Velocity operates as an embedded growth team: diagnosing bottlenecks across the entire funnel, building measurement infrastructure, and executing across acquisition, activation, and retention simultaneously.

Our methodology treats growth as a system, connecting paid media to CRM automation to product UX to analytics and pricing, rather than optimizing channels in isolation.

We serve startups (Series A through C), scale-ups, and enterprises across fintech, banking, SaaS, e-commerce, real estate, mobility, food tech, healthtech, logistics, insurance, retail, education, and travel.

Headquartered in Dubai with delivery capability across North America, Europe, and Asia-Pacific.

About the role

We are hiring an engineer to build those solutions with us, and for the foreseeable future, you are the engineering team rather than the first seat of one. You will set the architecture from the first commit with no legacy consensus to negotiate, and the two founders who currently write this code between client calls will be handing it to someone who does it better.

The breadth is unusual for a single role, because advertising platforms, analytics warehouses, CRMs, enrichment vendors and model providers all sit inside the same job, and you will see directly what the thing you built went on to earn.

The work is remote and full-time, with substantial overlap with Gulf business hours.

Please note that the compensation will be paid in USD, with no tax deductions applied.

What you will build

Every system here does something with real consequences, so expect to be working on:

Growth engines that run unattended: pipelines that source, enrich, score and act on data on a schedule, where a silent failure at two in the morning costs money rather than throwing an error someone notices.

Client-facing products: multi-tenant applications where one client's data must be structurally incapable of reaching another client's screen, with the isolation enforced in the database rather than in a conditional somewhere in the UI.

Deep integration work: most of what we build talks to somebody else's API, and that is where the hard problems are. Advertising platforms, analytics warehouses, CRMs, enrichment vendors, messaging and outreach platforms, document and spreadsheet APIs, and the model providers. Vendor documentation is frequently wrong, so verifying behaviour against the live API is part of the job rather than an exception.

The data and analytics layer: turning raw event exports into the answer to a commercial question. This is SQL and data modelling work, and it is a growing part of what we do. A meaningful share of the value we deliver comes from finding the thing in a client's own data that nobody has looked at.

Systems that spend money and send things: several of our pipelines buy credits, send email, and write to platforms we do not control. Those paths get designed to fail closed, because the cost of a wrong decision is not a stack trace; it is a client relationship.

You will also inherit and harden things that already run in production, and you will have real latitude to rebuild what deserves rebuilding.

How we work, and why it shapes whom we want

We build with AI models on every task, using Claude Code, Codex, Cursor, Kimi and whatever proves best next month, and you are expected to arrive fluent in this and to get further with it than we do.

That changes what we are hiring for. Producing code is no longer the constraint, so we are not looking for someone who types quickly. The constraint is that generated code fails in a particular way: it compiles, it reads correctly, the tests are green, and it is quietly wrong in a way that only shows up in production a week later.

We want someone whose instinct on being handed a working feature is to ask what would have to be true for it to be silently broken, and then go and prove it either way. The same instinct applies to claims, because we hold ourselves to stating only what we have measured and marking everything else as inferred or unknown, and that standard covers code, analysis and what you tell us is finished.

These are the shapes that survive a competent review, and catching them is the core skill of this role:

• A gate that blocks known-bad values instead of admitting only known-good ones

• A read that swallows its error and reports the failure as a zero

• A list endpoint that returns the first ten of fifteen records and looks complete

• A green build whose module cannot initialise in the runtime it ships to

What we are looking for

Strong production engineering across the stack: You have built and operated real systems end-to-end, rather than features inside an architecture somebody else owns. Node, with JavaScript and TypeScript both in play across our systems today, so we need you strong in one and fluent in the other. A modern React framework, Postgres, and the surrounding infrastructure of deploys, cron, queues, logging and alerting.

Genuine database depth: Schema design, migrations run against live data, row-level security and access control, query performance. Analytical SQL in a warehouse is a strong plus, and will become part of the role regardless.

API integration as a craft: You have been burned by an undocumented vendor behaviour and you have built the reading and verification habits that come from it.

Architecture that survives the second customer: Anything built for one client and monetised later gets rebuilt from scratch, and that is a lesson we have already paid for. We want multi-tenancy, configuration and isolation designed in from the first commit.

Ownership: You will often be the only person who understands a given system, so you write it to be understood, you know when it breaks before we do, and you say clearly when something is not finished. Both founders read code and will review yours, and we expect you to tell us when our review is not good enough.

Fluency with AI-assisted development, including where it is unreliable and what you do about that.

What earns you brownie points

None of the following is required, and any of it moves you up the list:

• Product sense, and views on what should be built rather than only on how to build it

• Comfort talking directly to a client or to a client's own engineering team

• Design and front-end craft beyond making the thing functional

• Experience with advertising, analytics or marketing platforms

• Data science and statistical rigour applied to commercial questions

• Experience raising the engineering bar around you, through review or mentoring