How to refuse a L+2 level task?

Your manager needs to be on your team to get promoted. Phrasing this as an ultimatum ("I'll only do this if I get promoted.") will almost certainly be counter-productive.

Instead, I'd phrase this as a collaborative discussion where you voice your concerns. Here are some starting points of what this could be:

When I look at the scope expectations for my level vs. the work I'm taking on, there seems to be a big gap. I'm concerned this may lead to problems when working with the broader team, and I'd love to chat with you about that.

I'm really excited to take on this project, but I'm also keen to understand how this fits into my path on the team. When do you think we could have a career conversation?

Rahul's advice is great. I also highly recommend my Effective Communication series to help with this: Alex's Guide To Effective Communication

Another thing to keep in mind is that there isn't really an inherent level of scope to projects. There is no such thing as an L5 project or an L6 project. How it actually works is that projects have a base-level and can go +/-1 (or even +/-2 if you're creative enough) depending on several factors like time, quality of execution, depth of system design, and more.

Let's take the time lever to an extreme to really illustrate this concept. Say you get a really complex project that feels super big - It has "Staff-level" (L6) project vibes. But let's say you get 10 years to do it: Any new-grad L3 can figure out things eventually, even if they don't know anything about the project space. So with the extra time, the L6 project becomes an L3 project.

So if they're not willing to budge on promotion, there are less nuclear levers you can pull besides "I just won't do this" - Namely de-scoping the project so it's closer to your level expectations. Here are some of those levers:

• Extending the timeline
• Splitting the work with other engineers, so you don't need to carry as much of it
• Lowering the amount of use-cases you need to cover
• Reducing the need for observability, so you don't need to add a ton of logging
• Not having a formal system design phase where you need to create a giant, complicated technical design doc and socialize it among the team
• Not super caring about edge-cases in general and being more amenable towards sacrificing code quality here and there
• Agreeing to ship it without an A/B test

For the future, the reverse is also true. Let's say you're an L5 (senior engineer) and you get an "L4-level" project. As a senior engineer, the onus on you is to sufficiently expand that project's scope so that it shows L5 behavior. You can do this by pretty much inverting the list I have above.

I talk more about truly understanding project scope here: "How to identify projects that are more suitable for senior engineers?"

Why? Why would you ever reject this opportunity? That sounds insane to me. You are effectively throwing away a golden ticket just because you weren't paid yet for the value of the golden ticket.

1. If you really were hired as an underleveled, then this project should take you less time than expected. That sounds like an easy "exceeds expectations" to me.

2. You can use this project to assemble a promo packet and get a promotion. Heck maybe 2 even. It's in your manager's interest to get you promoted so they look good too. He just needs the ammo from you.

I won't advise on how to do what you asked because its not an attitude I want other people to have. I'm not saying to work for free. I'm saying to take every chance you have to prove you deserve what you want. If it doesn't work out, at least you have projects beyond your level (L+2 according to you) and you can always go to a new job with it and get the title there.

I suggest not doing that though. Beyond L5, its very difficult to get a new promo by switching teams/companies. There's a lot of background knowledge that is built over time simply via exposure which is incredibly important for being "that guy" who everyone turns to during a fire. Take every chance you have to be important and learn something.

“To get what you want, you have to deserve what you want. The world is not yet a crazy enough place to reward a whole bunch of
undeserving people.”

―Charles T. Munger