Delivery
Sydney-led, Dhaka-built: how our hybrid delivery model works
Where our engineers sit, when you can reach them, who is accountable, and how quality is checked. The details behind the cost difference.
Razu Ahammed · Founder, Corexlab
4 min read
When people ask why Corexlab costs less than a Sydney consultancy doing similar work, the short answer is where our engineers sit. Corexlab is headquartered in Sydney, and our engineering team works from Dhaka, Bangladesh. We say that up front because it's the reason the numbers work, and because the questions it raises deserve straight answers.
This article covers how that model runs day to day: when you can reach us, who is accountable, how quality is checked, and who owns what.
Why we're set up this way
Senior engineering time is the biggest cost in any software project. Our Dhaka team is made up of experienced engineers whose cost base is well below Sydney's, and we pass most of that difference on. What we don't do is treat distance as a reason to lower the bar. The model only works if it gives you the same accountability, visibility and quality you'd expect from a team down the hall.
Working hours and the shared window
Dhaka is four hours behind Sydney (five during Australian daylight saving). Our engineers' working day overlaps the Australian afternoon, so there's a shared window every working day for stand-ups, reviews, decisions and pairing when it's useful.
In practice that gives you a useful rhythm:
- Your morning, you review what was built the previous afternoon and evening.
- The shared afternoon window, you and the team talk through questions, priorities and anything blocked.
- After it closes, the team keeps building on what was agreed, and the result is waiting for your review the next day.
Urgent production issues follow the support arrangement agreed for your project, not office hours.
One accountable lead
Every engagement has a single named lead, the person you go to about scope, progress, risk and quality. You don't need to manage a group of individual developers or work out who owns what. The lead runs the weekly rhythm with you:
- Priorities agreed at the start of each cycle.
- Small, reviewable changes landing throughout the week, not one large drop at the end.
- A written update every week: what shipped, what's next, and any decisions we need from you.
- A demo every week against the acceptance criteria we agreed in the spec.
If something is going wrong, the weekly update is where you hear about it first, from us.
How quality is checked
Distance makes people worry about quality, so we make ours visible. The same engineering practices apply to every project, and you can see all of them in your own repositories:
- Specs before code. Work starts from a written spec with testable acceptance criteria that you approve.
- Automated checks on every change: tests, type checks, linting and security scanning.
- Human review of every pull request by an engineer other than the author before it merges. We use AI coding agents to speed up the build stage, but agents never approve or merge their own work, and sensitive areas like authentication, payments and personal data get extra review.
- Small, reversible releases, demoed against the acceptance criteria and signed off by you.
We've written about this in more detail on our how we work page.
Who owns what
The custom code and deliverables we build for you are yours. Repositories live in your accounts from day one, not ours, so there's nothing to hand over at the end and you're never locked in. Third-party tools and open-source libraries keep their own licences.
Your contract is with Corexlab, the Australian company (ABN 81 668 487 189), headquartered in Ultimo, Sydney. We're happy to sign an NDA before you share anything sensitive, and we agree up front what data we can access and where it's processed.
Where the model fits, and where it doesn't
The hybrid model works best for product and AI engineering where the work can be specified, reviewed and demoed: building an MVP, taking a product from MVP to scale, shipping AI features, or adding a senior team to an existing roadmap. That's what our Embedded Engineering Pods are for: a senior team of two to five people working through your roadmap month to month, from A$25k a month.
It's a weaker fit if the work needs someone physically on site most days, or if your team needs to talk to engineers live during the Australian morning. If that's your situation, it's better to say so on the first call.
If you're weighing up a hybrid team against an onshore one, a scoping call is the fastest way to see how it would work for your product: who'd be on the team, what the shared hours look like for you, and what it would cost.
More insights
All insightsAI engineering
Evals before launch: how we take an LLM feature to production
A demo that works once is not a feature. How we build eval sets, regression gates and monitoring so an AI capability behaves the same way on day 90 as it did on day one.
6 min read
AI engineering
What a 2–3 week AI Opportunity Sprint actually delivers
What goes in, what comes out, and what it costs. A fixed-price way to find out whether one workflow is worth automating before you commit to a build.
5 min read
Working on something like this?
Tell us where things stand and what you need. We'll come back with questions and a suggested next step.