Tag Archives: startup class

Week 42

I’m sitting in a late night cafe with a piece of cake (no more coffee) and crashing. Trying to put the two weeks since the last update into terms of a progress progress report for this monday tradition of “horrible clarity”, and also making sense of things in the same time. Here’s an illustration how does it feel:

Messy desk with embedded computers and cables
Just a regular day debugging

Online learning

Sometimes still need to work on my fundamentals, and these days online courses seem to offer some benefits. They are not as good as mentorship, but as close as they can get since they follow my own schedule, and have a much bigger reach than the people I personally know around here.

The Startup Class is still going on, and this week’s two lectures were quite impactful.

  • Peter Thiel felt pretty much a waste of time – if you have done the assigned reading (otherwise it was decent). The reading was 3 of the chapters from his book, Zero to One. After the class I picked up the book, and finished by the weekend. I think it has a lot of gold nuggets that need to be dug out from the big pile of opinions and often strange conclusions. His ideas on monopolies and the destructive nature of competition is quite interesting and something I want to investigate more.
  • Alex Schultz is the VP of Growth at Facebook. His speaking style couldn’t have been further from Thiel’s, very active, engaging, challenging, practical. After 5 minutes I was a fan. In the end, though, I’m not sure what have I came away with, except for great speaker example, because the topics were beyond my experience and reach in scale.

There’s also a Successful Negotiation class on Coursera by the University of Michigan. The lecturer made a good case how useful it will be, and the first week was interesting, albeit probably the most amount of videos I’ve seen in a course, 2.5 hours for a week; though I was watching in 1.5x speed, that a comfortable pace somehow with most of these courses. In the end, most areas discussed (large scale negotiations and the US legal system) are again beyond my experience and reach…

Anyway, entrepreneurship porn is strong with this one


The much more focused way I’ve spent my time was getting on with the PCIeDuino electronics. I’ve been using Contextual Electronics’ Getting to Blinky video series to learn more of KiCad that I chose to use. It’s pretty good, I recommend it! After a few evenings of messing around I now have:

  • refined schematics
  • all components should have the correct footprint
  • the outline of the mini-PCIe card
  • the rat nest of components to place
Screenshots of the electronics design in KiCad
KiCad: component making, schematics, getting started on PCB

Now comes the next hard part: the actual layout. This should be another few evening’s surge this week, and after that hopefully I can start asking the knowledgeable people what I can improve on.


I’ve been figuring out whether to crowdsource this board, and given a few conditions it does look like a good idea:

  • Created & tested a working prototype
  • Have 2 months when I can have time each & every day to take care of the campaign
  • Have at least one more person who can help with the campaign (this is optional, but should tip the scale a lot)

If these are given, the next step is choosing how to actually do the campaign, e.g. what platform: HWTrek, Indiegogo, or something else? What target to set? What should be a good outcome and how to use the funds? (possibly external consulting, FCC certification, some other capital-intensive thing).

Coming week

Working on the side I’d be very happy with finished PCB layout, a concrete crowdfunding plan, and re-analyzed fundamentals based on things learned recently.

Be the first to hear about new scientific hardware development!

Week 40

One day goes after the next, and skipping on updates is not a good form, so here are some thoughts for this week related to Moonpunch, however small they might be…

Building a house in the neighbourhood in Taipei
Laying foundations of a house in the neighbourhood. Let’s make it a metaphor

This week’s notes

The Startup Class has started last week with their first videos, it was very highly anticipated by a lot of people. It’s focusing on creating hypergrowth Internet businesses, and it is not directly applicable to other styles (e.g. hardware startups), but there are some nuggets that I took home from the first two class:

  • The CEO’s number 1 job to manage their own psychology
  • Best predictor of success: every time the team has some progress to share
  • Vision/mission oriented ideas are the best ones: easier to rally around for the team and outsiders too, keeps people more on the path and going
  • Start at a very small segment and take all of it, grow from that – make something a small number of people love instead of something a lot of people like.
  • Sales fixes everything – when things are bad, making deals happen is the best way to turn things around, for the team as well

There are other series with such online videos/talks/interviews with inspiring people: Foundation is similar to the Startup Class in its selection, while I really appreciate Khan Academy’s Interviews with Entrepreneurs because it gives space for lot of other people, from the CEO of a large fashion brand to a founder of a small business focusing on university chemistry sets. These videos give an amazingly wide range of inspiration and examples of dealing with “situations” in business.

Also, I find myself having more hardware related discussions, and having more opportunities. Whether it is people coming to me as a (somewhat connected) member of the Taiwanese hardware scene, or one who at least took some steps towards making hardware (even if not enough steps yet), or as a commentator on hardware, it feels interesting, though not yet deserved. I don’t mind the opening of doors which it means, though, and hope to make the most of them in the way they fit the vision.

And yes, all of these does not matter at all. :) It’s so easy to lose focus with shiny objects like these above, while the only thing I need to figure out how to make a darn 4-layer PCB in KiCAD that I can send to be printed…