A microcontroller is a computer chip that is like a mini-computer. Your PC can do lots of things and is a 32-bit processor with lots of performance, memory and peripherals. An 8-bit microcontroller is a smaller version of this which solves smaller problems, has smaller amounts of memory and less sophisticated peripherals. The 8-bit microcontroller is ideal for adding "smarts" to everyday appliances like when your coffee machine turns on in the morning at a certain time. Another example, there are probably 10-20 8-bit microcontrollers in your car. One to control the power windows. Another to control the dimming in your lighting... get the picture?So.... basically I define what features will go onto new microcontrollers - how much performance, program memory and which interfaces are required for our customers target applications. I have to made decisions about trade-offs between features and the cost of the device. Microchip has hundreds of different microcontrollers - all with different amounts of memory, performance and with interfaces that will connect to things from USB (like on your PC) to Ethernet (for connection to the internet) to wireless protocols to something as simple as an analog-to-digital converter or serial interface. They range in price from $0.10 to $10 each depending on complexity.
You still awake? Well once I define a new product, I write a business plan to justify its creation. This includes analyzing the market for this product, analyzing the competitors, and doing a financial analysis to see if it is worth investing in the development of the proposed device. If the business plan is approved by all of the key players (marketing, design, applications, architecture, management) then we assign a team to design the product. I assist with the development by trying to represent the voice of the customer.
Once the device is designed, it has to be manufactured in a fab. Then we get back initial chips which I use for early adopter programs and to gain momentum in the market in anticipation of the product launch. I have to make sure all of the supporting collateral is in place - datasheets, press announcements, webpages, development tools, and then we officially launch the new device in the marketplace with a press release, training the worldwide sales force and often times we have a corresponding advertising campaign which runs in technical magazines.
At any given time I have a few products at various stages in this process. I interface mostly with the engineers and the sales team and act as a translator between them. But I also visit customers in order to understand their application requirements. I get about 60 emails a day which are questions from customers or sales people or engineers. I spend about 4 hours most days in meetings. In 99% of these meetings I am the only female. I speak a foreign language that involves lots of acronyms, NPSR reports, Baan flags, Stellent input screess, NPPS meetings, and pieces of technology called PPS, MSSP, SPI, I2C, and ZigBee. Who knew when I decided on electrical engineering to get that minority scholarhip money that the question "So what do you do?" would be so hard to answer??
So the new short answer will be: I am a marketing engineer, you really don't want to know any more. Let's talk about something more interesting and relevant to the world we live in because I already spent too much time thinking about work for the week!
2 comments:
Do you only define new products based on a customer's already expressed need, or do you define new products to try to sell them to a customer that doesn't know he/she wants it yet?
I'm not sure if my question is clear. I mean, does a customer call you and say, "I'm building a product that does XYZ and it requires a microcontroller that does ABC." Then you go define all the features and such? Or, do you define the features of a new microcontroller and THEN go out and try to sell it to customers?
And after you explain that, can you tell me what a transistor does? Just kidding!
Both - Certainly we take direct customer input into account, but we also try to anticipate the customer and market needs and define future-looking features they don't even know they need yet!
You've got the transistor thing down - or do you need another bar napkin tutorial??
Post a Comment