How much grunt work do you do?

  • Thread starter paralleltransport
  • Start date
  • Tags
    Work
I have to think about a problem and other times I just push the motions. Coming up with a good verification plan and testsuite is extremely challenging. In fact, verification of complex ASIC is one of the biggest challenges of design and one has to think about verification before even planning a complex chip. However, once the plan is made I can easily tell a smart AI what to do. That's the 80% remaining of the time where you are just pushing CAD tools and editing scripts. Most of that process is due to bad CAD tools in our industry (I can easily see that being automated).+1f
  • #1
Summary:: How much grunt work do you do?

Engineers out there,

What percentage of your time would you say qualifies as grunt work?

To clarify: Grunt work basically is work where you know exactly what the solution is (the problem is already solved), but you just need to push the motions mechanically to get things done. This could be paper work just filling forms, CAD work where you are just drafting something that can be done by a (reasonably smart) program, computer simulations to verify a design (if you're just pushing buttons and it could be automated by a smart program)... Even things like scheduling meetings, project management etc... is not grunt work since you are really solving a problem in that case that doesn't yet have a solution until you propose the optimal one.

I'd say between 70-80% of the work I do is grunt work. I spend maybe 20% of my time thinking how to solve a problem or learning more about it or planning, and 80% of the time just pushing the motions of verification or implementation.

I think undergraduate me really underestimated the percentage doing grunt work in my field. The reason being that all assignments in classes were challenging enough you spent most your time learning how to solve a new problem and exploring solutions. Once you figured out the trick to solve it, implementation was almost instantaneous.
 
  • #2
To clarify: Grunt work basically is work where you know exactly what the solution is (the problem is already solved), … computer simulations to verify a design (if you're just pushing buttons and it could be automated by a smart program)...
Design verification cannot be classified as something where you know exactly what the solution is. The design verification is an essential part of finding the solution. So I disagree that should fall under grunt work. Verification is the second most important step of the engineering design process (the first being properly stating the problem)
 
  • Like
Likes jasonRF, Klystron and berkeman
  • #3
Design verification cannot be classified as something where you know exactly what the solution is. The design verification is an essential part of finding the solution.
+1

Especially when you find issues or problems -- debugging and fixing the issues can be quite difficult at times. Some of the most complex problems that I've worked on (and thankfully solved) have come during design verification testing (DVT).
 
  • Like
Likes jasonRF, Klystron and Dale
  • #4
Design verification cannot be classified as something where you know exactly what the solution is. The design verification is an essential part of finding the solution. So I disagree that should fall under grunt work. Verification is the second most important step of the engineering design process (the first being properly stating the problem)

Coming up with a good verification plan and testsuite is extremely challenging. In fact, verification of complex ASIC is one of the biggest challenges of design and one has to think about verification before even planning a complex chip. However, once the plan is made I can easily tell a smart AI what to do. That's the 80% remaining of the time where you are just pushing CAD tools and editing scripts. Most of that process is due to bad CAD tools in our industry (I can easily see that being automated). You can think of it like this: if you were a verification manager you'd come up with a plan and tell your army of drones to implement it for you. Most of us don't have that luxury. Finally if you are an analog designer, the amount of checks you have to do in modern processes (5nm etc..) is absolutely insane, and simulation time is huge with parasitic netlists in the GB.
 
  • #5
Regardless of the mental effort required, if your definition of grunt work is “work where you know exactly what the solution is” then that cannot include verification. By definition you do not know what the solution is until after the verification. Often substantially after.

I think you may have a different internal idea of what “grunt work” is than what you described. It seems to be more based on mental effort than on the stage of development.
 
  • #6
Yes, I mean mental effort. Verification test plan development is essentially designing a test plan, which is mental work, while carrying out the plan (coding it up) is grunt because you know how to do it, just need to follow the steps of your favourite cad program (UVM etc...). Debugging failed test case is mental work.
 
  • #7
What percentage of your time would you say qualifies as grunt work?

To clarify: Grunt work basically is work where you know exactly what the solution is the mental effort is minimal
With the redefinition then I would say for me probably 50% - 70% grunt work would be a reasonable estimate. It is sporadic. Some weeks is close to 10% and some weeks is close to 90%.
 
  • Like
Likes paralleltransport
  • #8
The amount of grunt work I have will vary a lot from project to project, but overall I have been pretty lucky that it has been around 50%. I do have a PhD - folks at lower levels tend to have more grunt work than I do. It is one of the reasons why our better engineers often go back for advanced degrees.

My summer job on a farm during high school was different - probably 95% grunt work. 10+ hours a day. 7 days a week. I'll take my engineering job any day!

jason
 
  • Like
Likes CalcNerd, paralleltransport and berkeman
  • #9
Grunt work is some things and waste of time is other things; and then there is good sense and well planned and well conducted work. I want to comment but I should review the original question more carefully.
 
  • #10
@paralleltransport
A student in undergraduate courses may find that most or all laboratory exercises could be grunt-work. You should consider other situations, too. Scientific-based working positions which include laboratory, and working engineers who likely are working in a laboratory may have their own understanding of what is and is not grunt-work. Performing calculations on paper is not grunt-work; it is a necessary intellectual and technical skill or set of skills. If you need to take some measurement device into a field site and make some desired measurements, this can be a bit or more of labor, and you may call this as grunt work.
 
  • #11
Anyone who thinks making accurate measurements in the field is grunt work has never had to rely on measurements made by a grunt.
 
  • Like
  • Haha
  • Love
Likes Twigg, CalcNerd, Joshy and 4 others
  • #12
@paralleltransport
A student in undergraduate courses may find that most or all laboratory exercises could be grunt-work. You should consider other situations, too. Scientific-based working positions which include laboratory, and working engineers who likely are working in a laboratory may have their own understanding of what is and is not grunt-work. Performing calculations on paper is not grunt-work; it is a necessary intellectual and technical skill or set of skills. If you need to take some measurement device into a field site and make some desired measurements, this can be a bit or more of labor, and you may call this as grunt work.
I spend maybe <5% of my time doing analytical calculations, if at all at work. The best engineers in my field probably haven't done a diffEQ in ages, and there's some pretty high level of aversion to doing calculations... The level of mathematical illiteracy would surprise you.

Lucky for me I don't do much field work if any :). Lab is not a fun environment to be in, too noisy and uncomfortable. In school <25% of technical classes were lab classes for my degree, and they still involved analytical work.
 
Last edited:
  • #13
I'm in a simulation group we just use a fancy name like signal and power integrity. I hop between computer work mostly doing 3D EM simulation and lab work, and my typical day looks like below. I spend a lot of time thinking and staring at my desk real hard.

main-qimg-2ad8ac0503dad1a6bf175b6def20bad7.webp
 
  • Like
Likes gmax137 and Dale
  • #14
I'm in a simulation group we just use a fancy name like signal and power integrity. I hop between computer work mostly doing 3D EM simulation and lab work, and my typical day looks like below. I spend a lot of time thinking and staring at my desk real hard.

main-qimg-2ad8ac0503dad1a6bf175b6def20bad7.webp

Yeah that's exactly what the day is like for CAD intensive field, which is most of the semiconductor engineers I work with. In fact, a lot of the time we are debugging computer issues related to the simulator, not the design or simulation. I feel like this must CAD work must not be good for your brain.
 
  • #15
I think I've heard a punchline somewhere out there about physics and philosophy. The work for me seems very challenging and interesting, and very stimulating. I don't run into too many computer issues.
 
  • #16
I think I've heard a punchline somewhere out there about physics and philosophy. The work for me seems very challenging and interesting, and very stimulating. I don't run into too many computer issues.
Then you're a very lucky engineer. Circuit designers spend most of their time babysitting cad tools, especially when you go to very advanced nodes.
 
  • #17
Summary:: How much grunt work do you do?

Engineers out there,

What percentage of your time would you say qualifies as grunt work?

To clarify: Grunt work basically is work where you know exactly what the solution is (the problem is already solved), but you just need to push the motions mechanically to get things done. This could be paper work just filling forms, CAD work where you are just drafting something that can be done by a (reasonably smart) program, computer simulations to verify a design (if you're just pushing buttons and it could be automated by a smart program)... Even things like scheduling meetings, project management etc... is not grunt work since you are really solving a problem in that case that doesn't yet have a solution until you propose the optimal one.

I'd say between 70-80% of the work I do is grunt work. I spend maybe 20% of my time thinking how to solve a problem or learning more about it or planning, and 80% of the time just pushing the motions of verification or implementation.

I think undergraduate me really underestimated the percentage doing grunt work in my field. The reason being that all assignments in classes were challenging enough you spent most your time learning how to solve a new problem and exploring solutions. Once you figured out the trick to solve it, implementation was almost instantaneous.

Chemistry lab so my equivalent was 'pot washing.' Real grunt work.
Looking back it gave me a chance to pick up those things that was the difference between a uni lab and production/R&D lab.
It also taught me to be accurate, handling of chemicals, safety, different kit. Real consistency.
The nomenclature was old too although I was warned at A levels.
First week, I asked where the ethanoic acid was and the guys looked at me and said, 'acetic is over there.'
 

Suggested for: How much grunt work do you do?

Replies
6
Views
665
Replies
11
Views
2K
Replies
7
Views
769
Replies
30
Views
2K
Replies
6
Views
879
Replies
7
Views
2K
Replies
2
Views
2K
Replies
7
Views
1K
  • Poll
Replies
31
Views
3K
Back
Top