Announcing Airstone and the closure of Tinkercad – Updated

This post is no longer relevant! Tinkercad is alive and well, check the next post to read about Autodesk’s acquisition of Tinkercad.  We kept this post in the blog for posterity.

When Mikko and I founded our company in 2010, our core idea was to leverage a web browser UI and high performance computing to disrupt computer-aided design and engineering. Our vision was that a software platform created specifically for supercomputers would let us build some very exciting applications. In early 2011 we launched Tinkercad on this platform. It was the first cloud-based 3D CAD ever built and has grown to be a successful product in its category.

In parallel with Tinkercad, we continued development of the core platform. In October 2012, we launched a scripting interface for one of the key components, the Gen6 geometry modeling kernel. And finally, in late 2012, we had several major breakthroughs in our research work on the core platform that opened up application possibilities we had never imagined possible.

In response to these breakthroughs, I’m excited to announce an updated roadmap. There are two major parts to the new roadmap: 1) we are working on an innovative new simulation environment called Airstone, and 2) we will be discontinuing development of Tinkercad. You can read more about Airstone in the official announcement.

To help the engineering team focus on Airstone, I have made the decision to pull the team off Tinkercad. Aside from emergency maintenance, there will be no further feature development work and we will eventually turn the site off.

The Tinkercad shutdown timeline will be:

  • Effective immediately we have closed sign-ups for new users

  • August 31 2013 – All academic and free accounts will be changed to read only

  • December 31 2013 – All paid accounts will be changed to read only

  • June 31 2014 – Read only access for all users will be discontinued

There is an FAQ with additional details that we will keep updating as questions arise.

It was a hard decision for the team to stop working on Tinkercad. However, I can speak for everyone when I say we are very excited about the potential of Airstone. If you are interested in a sneak peak at the future of computational engineering, visit our website at or shoot us an email at

Yours sincerely,
Kai Backman
Founder & CEO


  1. I agree with nicolas,
    It’s really bad and brutal decision!
    My son Marvin (tollman), 11 years old just started 3 month ago and he like it and he learn a lot!
    Just keep it running without update!

    • We have been developing Airstone in parallel with Tinkercad for the past few months and after looking at the data there was still a lot of engineering time spent on maintaining the site. Tinkercad is a complex service with dozens of server types and hundreds of machines. There didn’t turn out to be an easy no-update path.

  2. Oh my god, I am really sad about this announcement. In my workshops I have trained dozens of users how to create 3D models easily with Tinkercad. Everyone loves it including me. I cannot remember how how often I recommended Tinkercad to all CAD newbies.

    Isn’t there a chance of a sort of Management Buyout to continue Tinkercad? I would be one of the investors.

    No matter how this story will continue, let me express my sincere thanks to the whole Tinkercad team and Kai and Mikko about this fantastic tool. It is really great and even product designers which are used to more complex CAD programs were fascinated about Tinkercad.

    Many thanks!

    Jochen aka

    • Thank you Jochen, your support has meant the world for us. I’m really sad to see the impact especially on children who have very few options in this space.

      We reviewed several options but didn’t find a satisfactory solution to keep the product going.

  3. As you said in FAQ the core system cannot be open sourced because it will be part of Airstone too. Is there a way to strip out the basic functionality of tinkercad from that core and let it run just for educational purpose maybe on a university server? So sad to loose such a great tool 😦

    • We thought a lot about this. However, the core system actually requires a cluster that is much larger than what most universities have. This also makes it hard for us to package it up as a desktop application, most people don’t run a high performance computing cluster.

      • I think alot of work is inside the client part of Tiinkercad – the editor with it’s great usability for example. Might it be possible to release this code to the public so that some programmers can build a own server backend to the client? Sure this might not get back the great performance and scale you have with the cluster now. Projects like Etherpad are hosted on many servers today to share the load without a central server cluster.

      • And yet… we all know there are many CAD programs, most of which have more functionality than Tinkercad, that run on garden-variety PC’s. It’s obvious that the supercomputer requirement stems not from the CAD functionality or UI, but from delivering those to thousands of users simultaneously via WebGL.

        I think what we’re all saying here is that we’d be Really Happy if some developer created a stand-alone desktop application that provided Tinkercad’s UI and functionality. That’s what I would have preferred (and been happy to pay for) in the first place, due to fear of what’s just happened, happening. I now have a year’s worth of work stored in a proprietary format on servers that are about to disappear. Hooray for cloud computing.

        Maybe Kai & Co. could sell or license some of Tinkercad’s guts to an interested developer and make a few bucks. Or take some of that new VC money and start a spinoff 🙂

      • Julia, there are a few more details to the puzzle. First, while a supercomputer isn’t required to run Tinkercad a large High Performance Computing cluster certainly is. The number of features in Tinkercad is obviously less than in a lot of CAD software but there are two features that require the power of Gen6: grouping and import. Grouping corresponds to boolean operations in most other applications and import is just that. Both of them are pretty hard problems to do at all, let alone fast.

        Airstone expands on the infrastructure that Gen6 relies on so it’s really hard for us to extract it as we need it going forward.

    • The main difference between Tinkercad and Etherpad is that only a fraction of the actual editing happens in the Tinkercad editor while for Etherpad it’s most of the work. The link between the client and the server is exceptionally tight. We did a large evaluation of available modeling kernels 12 months ago and didn’t find an alternative that worked. All the candidates we looked into crashed on over 10% of the models in the Tinkercad repository.

  4. This is just a terrible, crushing blow. Tinkercad has been a godsend to me and I use it constantly. I had no issue with paying a fee to use it, and if necessary I’d pay a higher fee to continue to use it. It has one of the most brilliant and intuitive UI’s ever. It has opened the door to CAD for so many people who couldn’t otherwise afford the time and or financial investment(s) required. I am devastated.

    • Thanks Julia, you have been one of our strongest supporters. We specifically wanted to keep the product running until December 2013 to make sure everyone has plenty of time to find an alternate solution. It saddens me to see the free users go earlier but the team does not have resources to support Tinkercad in addition to working on Airstone.

  5. Bummer! If this thing pays for itself, can’t someone take it over and keep it running? If you can’t keep it running, can we get another form of export for our existing models? I can see using OpenSCAD as an alternative, but exporting and importing loses all the internal geometry, providing only the final shape. I will have to re-create my things in OpenSCAD in order to get the ability to fine-tune objects. It would be a great help if we could salvage our work this way.

    • We are looking into providing a raw data dump of your files eventually. This wouldn’t import directly into OpenSCAD but it would make it possible to write an importer.

  6. Well Kai and team, thank you for all of the effort placed into this project. It made it easier to design the initial parts for my projects. Much easier to start off here and then import in to Cheetah 3D. The free version of SketchUp always disappointed me with its anomalies. I certainly could not afford their pro version for project. I hope that someone comes along and provides a similar product to Tinkercad and that your financial backers don’t throw down the litigation hammer upon them. I’m not sure what AirStone is going to provide, but it doesn’t seem likely that it will be a product I will need. But I’ll keep an eye out for it anyway. Thanks again.

  7. […] very unusual, given how regularly I used to write earlier. Today we all heard about the closure of TinkerCAD, the only other 3D modelling tool that works in browser like 3DTin. The silence on 3DTin’s […]

  8. Just curious as to what Airstone will be running. Electromagnetic simulations for electrical engineers? Finitie element stress analyses for mechanical engineers and rocket scientists?

      • I’ve done my share of EM simulations for RF/Microwave integrated circuit designs (all the passive structures). Normally you break up the problem into smaller parts so your individual simulations run anywhere between a few minutes and an hour or so, and then cascade the results later in the full circuit simulator. If you decide to let the simulators run overnight you might be able to fully EM simulate the passives of a complete functional circuit component like an amplifier or mixer in one shot, and that’s assuming EM code that leverages the 2 to 8 CPU cores on your standard engineer’s desktop computer. (Sometimes those software packages have diminishing improvement returns with more cores depending on how they are written.)

        Most circuit simulators have hooks to allow you to use the EM simulators from other companies using on your desktop computer or by linking to a UNIX box lurking somewhere in your facility. I could see this capability being leveraged by Airstone to run such simulations faster, even if not in “real” time. Of course this can open up a massive can of worms in terms of software licensing and by allowing proprietary design information to leave secure in-house computing environments.

        It seems like you might benefit from working with some of these larger companies to run and optimize their software to run on your computing platform remotely for their customers. With any luck you can can get some exclusivity agreement payments and eventual big buyout $$$!

        Now I must return to preparing for the Tinkerpocalypse.

      • Good analysis Pete. Our main challenge with existing software is that it doesn’t scale easily to thousands of CPU cores so it’s hard to achieve a decent speedup.

  9. I am wondering was that the VC Money that killed TinkerCad? Would it have been a viable business without VC investments?

    • I’m not sure I completely understood your question, please clarify in case my answer isn’t satisfying. The decision to close down Tinkercad was a management decision and ultimately my decision as a CEO. We would also not have been able to develop the product in the first place without outside investment.

      • Thanks for taking the time for the answer!

        I understand that without investments you couldn’t have developed TinkerCad into what it is today within such a short time frame. However investors want to see returns and in the fast paced Silicon Valley VC world you must produce exponential returns or at least exponential growth in traffic on your site.

        No one here assumes that TinkerCad is a failure as a business but I keep wondering if it was producing the kind of returns your investors envisioned when they put down their dollars.

        TinkerCad has no match in the CAD world. It touched on a market never addressed before and did it perfectly.

        The success (or failure) of TinkerCad is material to the 3D printing industry. The reason why DDD or SSYS fly so high on the stock market or MakerBot gets so much investments is because investors were lead to believe that the 3D printing market is expanding towards consumers from a professional market. Your decision goes completely contrary to this investment theses.

        Perhaps you have always thought Gen6 was your product and not TinkerCad. And TinkerCad was just a sandbox to perfect Gen6. It still bugs the question: Was TinkerCad a viable business only not growing fast enough to satisfy your investors?

  10. so Airstone is aimed at a completely different user than TinkerCAD?
    I see TinkerCAD as a makers dream, and a way for anyone that wants to just “sketch” out an idea. The times I have shown or taught it in a class the students lightup (young and old).
    Really think this is a big mistake. I would suggest talking with Google, MakerBot, or AutoCAD to have them take over the service.

    Very bad idea in closing it.

    • Thank you for your feedback. We explored several dozen options for keeping the product alive, including having a third party take over the service. We didn’t find a solution that would work with the development of Airstone. The Gen6 kernel is a crucial component of both products.

  11. […] This is massively disappointing, but totally understandable, as the company now focuses its development efforts on something called, “Airstone”. It’s an “interactive simulation environment powered by a supercomputer”. The idea is to provide massive engineering simulations for design work, based on technological discoveries they’ve made along the way.    We suspect the secret behind this move is simply money. How much money can you make providing free CAD services versus potentially pricey big-ass simulations for rich engineering companies? Yep. We’d do it too.    As for the loss to the 3D printing community, it’s not good. Tinkercad did provide a valuable service for many, but there are other options, including 3DTin and 123D Design. Will others emerge to take Tinkercad’s place?    Via Tinkercad […]

  12. As someone who develops in the Software as a Service (SaaS) space, you guys are doing a fine job with this in validating all of peoples’ fears about keeping their applications, yet alone their data in the cloud. Everything is there one minute and then – POOF! – gone. No install CD’s to pull out of the drawer. No Oracle database to go rescue data from using the local backup tapes. Nope. Just….gone.

    I think it would actually be more understandable if you just failed. The business model wasn’t sustainable and you just imploded financially and had nothing in reserve. No. This was a decision to turn things off because some greedy corporate VC’s shoved millions in your face and told you to. I stand by calling them greedy because…they are. That’s what they do. Lend money and expect massive returns on at least a percentage of their investments to make up for all the ones that FAIL because innovative technologists are no longer calling the shots but rather a boardroom full of suits.

    So, again a big thank you for burning bridges across the globe for everyone trying to innovate in SaaS or “the cloud”….which is just another buzzword-of-the-moment like the ones riddled through the board-approved Airstone press release talking about “disrupting” – really? Our sales guys moved on from “disruptive technologies” and “disrupting verticals” and all that nonsense years ago. The whole thing just reeks of biz speak and vapor ware.

    While my tone may be harsh, I think it is deserved. You have shamed yourselves by abruptly letting down every one of your users who counted on your efforts to learn a new trade, get excited about technology at an early age, pick up a new hobby or even start a new business venture….all those users with no recourse to another tool…no recourse to extract their data in a useful manner. You have also shamed yourselves AS engineers TO all your fellow engineers out there trying to convince people that this sort of service model works, is safe, is secure, makes them more productive, and saves them large upfront outlays by using a subscription model. I’m sure our sales folks are going to LOVE having to go pitch now when every client opportunity points at TinkerCAD and goes…well???

    • You’re darn tootin’ I will think twice and maybe thrice about ever investing time in a cloud-based solution again. I went with Tink because it it, well, so damned good that I put aside my distrust. One minute I was contemplating the gloomy prospect of having to learn some godawful user-hostile CAD program (I actually spent $300 on one and gave up on the thing) so I could make stuff with my new 3D printer; within 15 minutes of discovering Tink I was making stuff. That, and the fact that the people behind it seemed genuinely dedicated to the platform and concept, interacting with users and incorporating many suggestions from them. A huge letdown.

    • Thank you for your long and thoughtful post. I’d like to clarify a few things that you touch on. I’m strongly of the opinion that a controlled closedown is better for the users and I want to emphasize that the decision was management led and all of management is engineers. Ultimately it was my decision as a CEO and I stand fully behind it.

      First, I’m strongly of the opinion that a 1.5 year staged closedown plan that Tinkercad is executing is immensely better than going out of business without warning. In my view it would be completely irresponsible to just let the money run out and shut the site down. In addition to working on our own raw export tools to let everyone migrate out as much data as possible we are strongly encouraging and co-operating with entities like the Internet Archive to scrape our pages to extract all public content available.

      Second, as CEO there are sometimes business decisions that require the company to move out from one line of business to another. As an engineer I feel it would be completely irresponsible to not act on those decisions. Dithering on a decision to do a required close down of one line of business would make it impossible to do a controlled schedule like we did with Tinkercad.

      Finally, as a team we are really excited about Airstone. We know how important Tinkercad was and we would not have made the decision to move on unless the were firmly of the opinion that Airstone would be more important in the long run. Posterity will obviously judge us by what we deliver which is why we decided to focus fully on developing Airstone.

  13. Noooooooo!!! You can’t give up! Tinkercad is just about the coolest program I know of! You should see my 6-year old designing snowmen Christmas ornaments! It’s so simple, he and I did the tutorials and he went right to work!

    This new multi-fancy simulation thingie should cool and all but is it going to be keep my 6-year old designing (and thus building) and eventually engineering amazing Thurgs?

    And all that code…. you’ve honestly built a beautiful, elegant online design environment. What will happen to that code? The UI editor … so preeeety…

  14. Can you point us to the next-easiest 3D CAD program? In your opinion? Or name a few? Are there any that are similar to TinkerCAD? The idea of using common shapes to build up sequentially more complicated objects?

  15. Before I say anything negative, I want to say this: I appreciate that you have always been communicative both with personal emails and in the community. You have been outstanding with that, unlike many companies.

    However, your spin aside, the reality is this.
    I have seen this many times over the years. You overbuilt. That is, you built software that you could not afford to support with your business model. Your job as CEO was to build software that could be supported with the revenues of the Tinkercad business model, or build a business model that could profitably support your software. You failed at that, sorry.

    At some point this became obvious to you as a manager and an engineer. It happens in software businesses sometimes. So you looked at alternatives to protect your main asset, the software engine, and your new venture is the result. I am happy you were able to apply your engine to that.

    If you recall, I sent you an email questioning your Tinkercad business model when you announced your pricing, which I considered too high for the market. I now see that you did this higher pricing model as a means to survive while raising venture capital, or you mistakenly believed that enough people would pay for the Tinkercad software at those high prices to support your overbuilt software. This was a solid business decision.

    Now you have raised some venture capital, and that took some time to do. So you have known for some time that this was going to happen. So you took new subscribers at a time when you knew they would not be able to utilize your software model, or have editable models in the future.

    Except, what did your recent new paying customers buy when they bought your service? That is, what did you sell them? You sold them the ability to develop models, but you also sold them a service that allowed them to collaboratively share models, make changes to others’ models, and store their models online. You sold them the ability to develop models that they might change over some period of time. Your top tier model allowed access to develop thousands of models. Your top tier purchasers were expecting to have something more than just the ability to log in, put up a quick model, and then print it. They fully expected their models to be available to them in the Tinkercad format They were buying a cloud service of models in a proprietary format with the expectation they would be available for modification in the future. They invested time and money in their expectations. Likewise, it appears you had some level of services you promised too other customers such as engineering groups and educational institutions.

    So you took their money, and you are not delivering that service. And you are taking the main asset of Tinkercad to another firm. Unfortunately, that is not a fair business practice.

    In most businesses, when you take peoples money and don’t deliver, those people who paid for the service have access to the assets of the business as a means to recover their investments. You are not allowed to just take those assets and wander off to another venture with them. This is the price a business person pays when taking customer’s money and not fully delivering the service that was sold to them.

    I am speaking for those who did pay and might not have thought this situation through.

    My questions would be:
    When I paid for the Tinkercad service, what was Tinkercad selling me? Did I get what I paid for?

    If not, what is due a paying customer? A refund? Editable versions of models they have designed with the service? Restitution for time and money for planning and expenses for plans I made around the expectation that this paid service would be available? Normally, customers have a means, via bankruptcy or reorganization, to either get some of their money back or access to the assets of the company.

    Sorry Kai, you failed at your Tinkercad business. But is it really fair to say you can’t offer the product open source because it would hinder your future venture? Is it fair to say that you can’t offer a format that customers can utilize to convert their existing models to an editable format because you just are “no longer supporting it?” It is fair to have taken their money to help support your development on an engine that you intended to take elsewhere when you knew you were doing so quite some time ago. Is it legal?

    Business does not work that way. You took money. There was an expectation of ongoing services. Your business failed. You owe something to at least the paying customers as a result of this. If you do not address these issues directly, I could see those customers getting together and pushing you into bankruptcy. If I were a prospective venture capital firm, I would want these answers before I put money into the new venture.

    Kai, you know I like your team. And again, I appreciate you have always been open. But you really need to figure some stuff out for at least your paying customers.

    • Thank you for your long and thoughtful comment Perry. I’ll focus on what we are concretely offering to paying customers and some analysis behind it. But before that I want to touch on accountability. Given I’ve been CEO since the inception of the company I want to re-iterate that the decision to close Tinkercad is fully and completely my responsibility.

      That said, there are three things we are doing right now for the paying customer of Tinkercad:

      – We refund any purchases done in the past 60 days
      – We offer the service to paying customer for 9 more months
      – We are working on a raw exporter that will dump as much as possible of your Tinkercad design data

      These were all covered by the FAQ with one exception. We have been advertising a 30 day refund from purchase date but our internal policy has actually been 60 days. I’ve amended the FAQ accordingly.

      The first and obvious measure is the refunding of purchases. We have always liberally refunded any purchases done but I think it’s important to emphasize it in this context. Benchmarked against other organizations I think we are doing a decent job with addressing the concerns of users who joined recently.

      The second measure is continuing to offer the service for 9 more months. This is obviously above and beyond any expectations set by the Terms of Service. However, I also think this is a reasonable period of transition time for a service that was billed month to month.

      Finally we are working on a raw exporter for the design data in Tinkercad. The exported data wont be directly compatible with other tools but we aim to have a textual format that is easy to parse and that includes as much information as possible (full design tree, full revision history etc). I strongly believe that users have the right to their full data and this is our way of addressing that.

      • Thanks for the info. I appreciate it. I don’t see how you can do much more than that. Above expectations, I must say. With that, I wish you good luck in the future. Let me know if I can be of any service to you or your team in their future endeavors.

  16. Thank you for creating Tinkercad! Even though I respect your decision to shut the service down, I can’t help but feel a bit disappointed. 2 thoughts:

    1. You’re leaving an interesting niche of the CAD/3d printing market in which you were the only real competitor
    2. I think a lot of the value of Tinkercad is actually in the UI design and workflow, not in the Gen6 engine

    Both of these probably influenced your decision to move on to Airstone (bigger market, gen6 becomes key) but I’m curious if this was mainly driven by the need for bigger proftits or the engineering challenge?

    Now that I can no longer easily create 3d models for my printer, I’m going to start working on my own simple CAD editor. I really believe there is a need for one 🙂

  17. Damn! Tinkercad was useful. Very sorry to see it go, maybe the idea was just a little bit ahead of the general market, hopefully it will make a reappearance in a few years.

  18. any thoughts on allowing a export format to allow us to more easily adopt a replacement product? If you are going to cut us off may as well make it easy to transfer designs

  19. […] Tinkercad era una de esas aplicaciones que sus usuarios adoraban: modelado 3D sencillo pero potente, gratuito, desde el navegador, con una comunidad de usuarios muy activa. Y de repente, el pasado 26 de marzo, anunció su cierre y el trasvase de todos sus recursos a un nuevo proyecto con un enfoque diferente …. […]

  20. […] Tinkercad era una de esas aplicaciones que sus usuarios adoraban: modelado 3D sencillo pero potente, gratuito, desde el navegador, con una comunidad de usuarios muy activa. Y de repente, el pasado 26 de marzo, anunció su cierre y el trasvase de todos sus recursos a un nuevo proyecto con un enfoque diferente …. […]

  21. […] to learn more, but couldn't. Maybe he can post a link to that info for me. Try this links: PANIC! RELIEF! The shutdown plan has been rolled back and effective immediately new users are again […]

Tell us what you Tink

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s