An in-house team delivers the most control. The engineers internalise the business domain over months and years, and this context remains with you. The price shows up as slow hiring and fixed overhead: recruiting a strong engineer is slow, ramping up adds more time, and the salary keeps running whether the roadmap is full or empty.
Full outsourcing means someone else is accountable for shipping: they staff the team, the partner manages the process, and the provider carries the risk of missing the date. This works well when the work is a defined project and you have someone who can make decisions quickly. It breaks down when the requirements change weekly, because a vendor will not guess what the business wants.
Team extension falls in the middle: you bring in developers but keep the planning and the management in-house. It moves quickly — a suitable engineer is often available far sooner than a new hire developers in london — and the commitment ends when the work does. The catch remains that your technical leaders must have the bandwidth to manage them. Without strong internal leadership, the result is paying for hours, not results.
Most of the time, companies blend them. One durable pattern keeps architecture, product decisions and core domain code inside the company, while an outside vendor handles discrete features, migrations or mobile clients. The principle is simple enough: keep the parts that are hard to re-learn, and contract out the well-trodden work.
A few questions generally decide the matter. To begin with: is this typescript software central to how you make money, or a supporting tool? Then: over what horizon does the work continue — months or years? Third: who will maintain it outsourcing company in two years? Answer these three honestly and the model usually chooses itself.