Where can i find 2020 Census Data at ZTCA or ZipCode

You probably already know this but.. do keep in mind that zip codes are created to deliver mail and not really a great geography for doing data analysis. If you have the option, I'd suggest using Census geography like tracts or block groups. 

here's a nice article from a few years back:
towardsdatascience.com/stop-using-zip-codes-for-geospatial-analysis-ceacb6e80c38

While it is true that census geographies are preferred by technical users, it appears that for presentation to popular audiences zipcodes are immediately recognizable and understood. However, one should bear in mind the difference in geography sizes with c. 46k zip codes versus 73K census tracts. One option to deal with this difference is to give information using ZCTAs but use GIS to show a thematic map within the ZCTA showing either tracts or block groups - highlighting differences within the ZCTA.

I find that most people who don't do the kind of work we do are not familiar with most geographies which aren't states, counties or municipalities, so using those as references often makes sense. It also makes sense to incorporate whichever census geographies are relevant to the discussion. I wouldn't add a Zip based geography unless it was necessary. Landmarks are better cues than zip codes most of the time.

I agree with David.: only use ZIP or ZCTA as a last resort.
And other pro-tips: If what one is publishing or presenting can include graphics -- then include maps! And maps of block groups will look so much nicer than maps of ZIP codes (which are bizarrely shaped). 
On the other hand, if what one wants to publish has no maps -- but is a list, or some kind of look-up tool -- then the geographic units would have to be what the audience knows. So I often prepare tables with cities or towns named. 
If the report or web resource was for somewhere with a large city (a city that is >100 square miles), then I recommend one "roll up" and summarize data into sub-city neighborhoods -- again mindful of what the audience knows.
A great example here in Mpls-St Paul is the Wilder Foundation's MNCompass project. MNCompass regularly assembles socioec data into neighborhoods or districts. www.mncompass.org/.../minneapolis-saint-paul
This gets good audience response because (again) people mostly know what neighborhood they live in -- even if they don't know their census tract number. 
--Todd Graham
  Metropolitan Council Research 

I agree with David.: only use ZIP or ZCTA as a last resort.
And other pro-tips: If what one is publishing or presenting can include graphics -- then include maps! And maps of block groups will look so much nicer than maps of ZIP codes (which are bizarrely shaped). 
On the other hand, if what one wants to publish has no maps -- but is a list, or some kind of look-up tool -- then the geographic units would have to be what the audience knows. So I often prepare tables with cities or towns named. 
If the report or web resource was for somewhere with a large city (a city that is >100 square miles), then I recommend one "roll up" and summarize data into sub-city neighborhoods -- again mindful of what the audience knows.
A great example here in Mpls-St Paul is the Wilder Foundation's MNCompass project. MNCompass regularly assembles socioec data into neighborhoods or districts. www.mncompass.org/.../minneapolis-saint-paul
This gets good audience response because (again) people mostly know what neighborhood they live in -- even if they don't know their census tract number. 
--Todd Graham
  Metropolitan Council Research