I've seen this a lot at Intel during my time here (less than 10 years). Recent examples are 2 task forces , one where the "issue" was proven by modelling to not cause impact to performance but the lead says "if we don't fix it we look bad". Second instance , a minor issue was found that can't even be detected by metrology, was noticed recently due to a secondary activity, and looking back in historical data it was always there. Daily task force ensues to fix the problem. The Lead says "even if it is proven to be low risk the "defect" needs to be eliminate. Is intel overdoing it ? Do other companies (apple, qualcomm, nvidia, amd, etc) do the same or would chalk it up as low risk and move on ? My impression is that we are wasting hundreds of engineering and tool hours on issue that doesn't impact customer at all. #semiconductor #manufacturing
I see. That makes sense. On my side i have been trying to see how to get more on decision making and less on execution (like up the food chain). Current org is seen more as support/partner than strong influencer. Planning to move soon internally, but started to get tempted by external after a couple colleagues left for greener pa$ture$ lol
Where did they go for more money?
Downstream issues are sometimes hard to translate into yield or inline metrics as there is no/low visibility. That doesn't mean the issues don't impact bottom line. U should talk to someone u trust to bounce ideas on examples of those types of issues
India
4h
1743
Why is it so G*damn difficult to move money out of India
Tech Industry
Yesterday
879
East Asian Men don’t talk to me bc I’m Vietnamese
India
Yesterday
1112
What do vegetarian Indians eat for protein?
AMA
Yesterday
1614
I have worked at TikTok US core tech for 3 years. AMA.
Health & Wellness
2h
393
Issues with sleep
Just because metrology doesn't find it doesn't mean it won't impact yield down the line. Ask me how I know :) I suspect your Integrator/Defmet is either an idiot or brilliant.
Definitely brilliant :) these teams are really good. But it's in this case no yield impact. Hence my concern . Is it worth it ? But maybe I'm oversimplifying things :)
You have a fair point. I'd try to dig into it little by little if there's a mountain of other motivation behind the decision. Unfortunately experienced folks sometimes have no time to thoroughly explain but do appreciate if you come to them asking. Then again experienced folks wouldn't say "because we'll look bad"