Bob Dubner

My name is Bob Dubner. I am an engineer.

Engineer:  A person who solves problems you didn't know you had using methods you don't understand.

Specifically, I am a practitioner of the gentle art of getting computers to do what we want them to do.

I was one of those kids who liked taking things apart because I wanted to know how they worked. I was lucky to have an electrical engineer father who dove deeply into computers starting around 1958. (I was five.) When I was about ten I asked him what electrical engineers did, so he brought home some relays and batteries and some simple electronics. When I was about fifteen, I asked him what computers were; that time he brought home a self-instruction course in FORTRAN.

He did, first, explain that a computer was basically an adding machine, but with the ability to do a calculation and then make a decision about what to do next based on the result. It was a long time before I realized that was a remarkably succinct description of a Turing machine.

I learned FORTRAN quickly. Within months I was earning money writing an inventory control program for a local businessman who was reluctant to pay $1,500 for my dad's company's professional services; but he let himself be talked into letting a fifteen-year old kid take a stab at it for $300. (They used that program for about twenty years.)

In 1970 I got a summer job at NASA's Goddard Institute for Space Studies in Manhattan, programming the monstrous IBM 360/95.

And that launched the journey for me. I started with FORTRAN on the supercomputers of the time, but quickly ducked back down to assembly language on minicomputers and the brand-new microcomputers. While getting an electrical engineering degree (1971 through 1975 at MIT) I worked summers with the Intel 4004, then the 8008, and then the 8080. When the 8080 proved too slow for certain jobs, we added in a bit-slice RISC processor. In the 1980s I started using IBM PCs. Soon after that I was neck deep in the revolution fueled by competition between Intel's evolving x86 architecture versus Motorola's 68k hardware; I designed computers based on both architectures.

I got to do a lot of neat stuff.

If you look at my résumé, you'll see that it's currently top-heavy with my recent experience in the financial world. That is nothing to sneeze at: it involves data science; database design and development; mastery of SQL; development and regression testing of mathematical financial models; experience with mathematical optimization methods; the practical necessity of coping with massive amounts of incoming daily data from multiple sources ranging from beautifully implemented XML on the one hand, to scraping data from PDF files on the other.

But that impression wouldn't include some experiences that are my favorites. Using strong cryptography to protect the guts of casino slot machines against hackers.  Building tools for videotape editing.  Designing multiple generations of video production graphics and character generators. Building an evolutionary series of software and hardware machines for rapidly manipulating numbers with hundreds of thousands of digits for number theory research. Designing a special purpose microcomputer, back in the mid-1970s, that went into a research automobile to shift the gears and operate a radar-based collision mitigation and cruise control system. 

Although I am currently gainfully employed, new challenges are always worth considering. If you would like to talk to somebody who thrives on solving tough problems, please, get in touch. You can reach me at rdubner@dubner.com