You design a data ingestion framework by first understanding the requirements and then the goals of the project/company. After that, you take a look at the Company technology portfolio, see If there are some tools or processes already homologated for these requirements and goals, or if you need to bring something new to the table. Some questions you can make are "does latency is expected?" "Is it tolerated?" "Or we need this data as quickly as possible". You make a decision tree based on these requirements. This decision includes not only ingestion patterns, but also data store patterns. You need of course to also analyze costs from day 0. If it is something that might be a problem then you must be aware since beginning. This is the part that highlights High skilled people, because this is where you show that you are able to do more with less, finding solutions that nobody finds.