Transitioning from a full-time, paid position to consulting, part-time or teaching takes longer than you think it should. I spent my entire career trying to do things my employers could not take away from me. When it was time to transition from a FT position to a combination of smaller efforts it was time-consuming, hectic, disappointing, and ultimately extremely satisfying.

My network of colleagues was a bust. Most were just good luck, I am sure you’ll find something. Some of those did come through with great part-time work and great people to work with but I needed to invest in my skills and engineering tools of the trade.

I spent an enormous amount of money retooling (hardware, software, paid training, personal research) to do up to date engineering. All those years as a manager and executive were left behind.

When it is time for a transition, be prepared. It will take six to twelve months and tens of thousands of dollars to really get ready. Think ahead. Just like your 401K and other retirement plans, start early and have a plan.

Good engineers just can’t go from 100 mph to 0 and be happy.

I am all in with MATLAB. I have my own license and toolboxes that I use for consulting and teaching. I do constantly check out the free alternatives but always stick with MATLAB. MATLAB does have a premium price but it works together, works with my acoustic data collection system, and now with my wifi data collection sensors connected with Thingspeak. Plus, with a license you receive quick help. I told them recently there was a problem with their system identification function. They asked me to send the transfer function data, which I did. They thanked me and agreed there was an issue.

With the free alternatives I can’t get past “it’s almost as good as MATLAB” and “well, the tools are about 15 years behind MATLAB at this point.”

Project management can be a thankless job. You have all the responsibility and little direct control of resources. Like many things in life, sometimes what you do early in a challenge can make a big difference on how things turn out later. Building on the sayings that you cannot fix design problems in manufacturing and the first amplifier sets the noise figure in a receiver chain, project management becomes much easier if you put your best resources on the job upfront.

I have a PhD friend who is a patent agent working for law firms. He mentioned he spends a lot of time telling applicants their data is bad. Data validation, especially hard to get even once data, is incredibly important. Doing noise investigation work in extreme environments I had a primary digital recording system, a backup digital system, and even a video camera with good audio recording capabilities. I used to buy inexpensive color printers to print out analyzed data onsite. Always come back with the data you are after.

When candidates come to interview with me, they have checked me out in LinkedIn and have done a google search. It is surprising that when I talk with someone new, especially a senior-level accomplished person, I discover that they have no LinkedIn page and practically no digital footprint that google can uncover. It is easy to create a profile that you control and it is fast to update.

Sometimes you want to know if a new process or technology insertion is not only effective, but how effective with recommendations for positive change. Statistics, now matter how powerful, may be only part of the answer. Borrowing from the social sciences, evaluation research is a methodology to measure improvements.  

It is here to stay, just like the internet and color photography. This modality is not a panacea. There are numerous examples of really good ones and not so good ones. The latter usually quickly put together and deployed to make a quick buck. The good ones today might have started out not so good, but improved with the application of reasonable learning objectives and up-to-date tools, methodologies, and standards.

Never say you have seen or heard it all.