Delft borders The Hague, about ten minutes away. TU Delft is a strong source of tech startups, and a number of our clients are TU Delft alumni.
Book an intro →
NedDev builds websites, SaaS platforms and AI systems for companies in Delft, from our studio at Den Haag in Rotterdam. Delft is nearby, with a travel time of fifteen minutes. The city runs on TU Delft and the YES!Delft incubator climate, with many deep-tech and engineering startups. A fair share of our clients are TU Delft alumni. For that market we build technically strong, scalable software. Short lines, regular time at the table. Projects start from €2,500.
Web development in Delft means building web applications, SaaS products and AI systems for a city entirely focused on technology. TU Delft is an engine for deep-tech, engineering and hardware startups, and incubators like YES!Delft make sure research grows into companies. Those teams are technically strong and look for a build partner who thinks along at the same level.
NedDev works as a The Hague studio for clients in Delft, and Delft is so close that we are practically neighbors. Some of our clients studied at TU Delft, and we speak the language of engineers: precision, reproducibility and architecture that holds up.
For companies in Delft we typically build:
Delft is perhaps the most technical city in the Netherlands. Around TU Delft and YES!Delft, startups emerge that come out of serious research: robotics, energy, materials, sensors. The founders are often engineers who know exactly what they want and who value precision. Anyone building for this market has to deliver software that not only works, but is also reproducible and verifiable.
That is exactly how we build. For deep-tech startups we use our multi-tenant SaaS architecture, so a product can grow from the first customer without a rebuild. For applications with AI we value determinism: in AI development we replace the guess of a model with a fixed calculation where possible, so the same input always produces the same result.
Our technology is complex, do you actually understand what it is about? We do not build the physics, but we do build the software around it, and we are glad to dive into the details. For JinSulate we built a system that reads construction drawings geometrically, so we are used to technically substantive projects.
The City of Delft positions itself as a technology city, see delft.nl. How we secure reproducibility in a technical product is shown by JinSulate, where a geometric counter replaced the loose AI estimates so results always come back reliably.
Delft is about a fifteen-minute drive from our studio in The Hague. That is so close that sitting at the table is never a hassle. We come over for the kick-off, for design sessions and for the go-live, and we can switch gears quickly in between if a topic calls for it. For the intensive early phase of a deep-tech product, where many choices are made, that proximity is valuable.
We build our frontends on Next.js, for fast and easily findable interfaces, with Laravel or FastAPI on the back end. That consistent stack ensures a product stays manageable even after delivery, which matters for a growing startup.
A Delft project usually runs like this:
Many TU Delft alumni register their first company at the Chamber of Commerce and then look for a build partner who can handle their ambition. For such teams a well-defined first platform is the smartest start, and the short distance makes close collaboration easy.
Our internal stack packages for multi-tenant SaaS. A Laravel + Filament starter, an audit-trail engine, and a tenant-impersonation package that runs across 12 clients.
8 PACKAGES · 2.4K STARS EDITORIAL · LONG-FORM ↗What we write down as we learn it. Case studies, technical write-ups, design decisions. No content marketing, just real knowledge.
42 ESSAYS · MONTHLY RESEARCH · AI EXPERIMENTS ↗Side projects and R&D. Voice-agent prototypes, RAG pipelines, AI knowledge-graph experiments. Some become products. The rest teach us something.
14 EXPERIMENTS · LIVE DEMOSYes, that is one of the groups our way of working suits best. Around TU Delft and YES!Delft, startups emerge from serious research, with founders who are often engineers themselves and who value precision. Those teams look for a build partner who thinks along at their level about architecture and reproducibility. We build software that not only works but is also verifiable, and for AI applications we replace the guess of a model with a fixed calculation where possible. For JinSulate we built a system that reads construction drawings geometrically, so technically substantive projects do not scare us off. We do not build the physics, but we do build the reliable software around it, and we are glad to dive into the details.
Delft is about a fifteen-minute drive from our studio in The Hague, so sitting at the table is never a hassle here. We come over for the kick-off, for design sessions and for the go-live, and we can switch gears quickly in between when a topic calls for it. Especially in the intensive early phase of a deep-tech product, where many technical choices are made, that proximity is valuable. The ongoing work happens digitally, with a fixed weekly demo where you watch the product grow, so we lose no time on trips that could just as well have been a video call. Thanks to the short distance you get intensive guidance without it burdening the schedule.
Projects at NedDev start from €2,500, but a full SaaS product or AI system costs more. SaaS involves user management, scalability and ongoing maintenance, and with AI the cost of the model and the control around it are added. For deep-tech clients reproducibility weighs heavily, and we build that in with logging and fixed calculations where we can. We would rather give you an honest range after a short conversation than a number that looks too good up front. The ongoing costs for hosting, management and any AI usage we make clear in advance and monitor with hard limits, so a spike in usage never leads to a surprise on the bill.