Engineering Tools Are Broken

The problems (yes, there are multiple)

As I write this there are entire companies built on MATLAB and Excel. That’s a bit funny because Excel wasn’t ever built for engineering and MATLAB was released in 1984 (what other 40 year old software do you use?). When you think about it, it really seems that we should be able to do better than an accounting tool and decades old, pre-internet software. The engineering discipline is the foundation of our society, yet current engineering tools are clearly suboptimal. The problem is (at least) fourfold:

First of all, highly trained engineers are spending their time on tedious and boring tasks like plotting results, parsing data and figuring out which of the brackets in their equations isn’t placed right. This may sound a bit banal, but a lot of time is wasted on this. And that’s time that isn’t spent on what really matters (and frankly, what’s more fun): building the future.

Secondly, in engineering solutions aren’t shared to nearly the degree that it’s done in software. When I write software, it’s easy for me to use someone else’s solution to a problem. It’s plug and play. And I can easily see whether it’s a reliable solution too. There’s versioning, Github Issues, you can see how active a given repo is and how many people are maintaining it. In engineering, this doesn’t happen. What tends to happen is that engineers Google around for MATLAB scripts, copy-paste them and then sort of mess around until they work for their use case. In other words, engineers are implementing the same solutions over and over rather than sharing them. This doesn’t just mean they’re spending time doing unnecessary work, but it also means that a given solution might not be implemented optimally. Now, there’s an obvious reason why engineers aren’t sharing their solutions: there’s no incentive for an engineering company to do so. Why randomly give away your IP? But what if we could incentivize them to do so? What if we could create a culture of sharing and re-using solutions like what we have in the software world?

Third, it’s hard to collaborate with current tools. Part of the reason is that Excel doesn’t properly support versioning (Git), but the bigger issue is how fragmented the ecosystem is. One engineer will be using MATLAB, another will prefer Excel, and yet another will want to work in Python. This makes working together extremely hard, and in practice it means that “collaboration” happens by emailing files back and forth, or at best through a shared Excel file containing everyone’s latest calculation results.

Fourth, there’s performance. This one’s an easy fix, but seriously: why do I need to download gigabytes of ancient software to run locally on my machine? Why not run your code on a supercomputer somewhere and engineer from your browser?

Okay, I know I said I’d list four problems, but there’s one more thing I’d like to mention and that’s the look and feel of current engineering tools. Imagine you’re an engineering student. It’s your first day at university. You’re ready to build the future. To build robots, nuclear fusion and space habitats and then you see… this:

Does this look like the future to you?

Does this make you feel like you’re going to build the future? Let’s maybe try to make things a bit more J.A.R.V.I.S. and a bit less Windows 95.

Our solution

You didn’t think I was just going to list problems, did you? Luminal is here to solve all of these issues. If this sounds interesting you can request early access here. We’re going to be running a (very) small closed beta because we want to make sure we get it right before releasing anything to the wider community, so the earlier you register the earlier you’ll be able to try Luminal out.

Speaking of community, Luminal isn’t meant as * just * an engineering tool. It’s meant as a community too. We can’t (and don’t want to) do this alone so our goal is to bring together as many engineers as possible. Anyone interested in making engineering better can join. And you can help us make Luminal the ultimate engineering framework by joining the discussion and letting us know what we can do to improve your engineering process.

Join the discussion here.

Some final words

This is obviously just the beginning, and we have no illusions about how difficult this will be, but we’re extremely excited to start this and help make engineering better. We have some even bigger plans that we’re not ready to share just yet, but suffice it to say that we’re not just thinking software. If that doesn’t get you excited, I don’t know what will!

Co-Founder @ Luminal