Even when I was in large companies with over 10K employees, there were times due to internal politics I could not interact with more experienced folks. Truth be told, the difference between fresh graduates and seasoned people... is the more experienced people learned from the school of hard knocks.
Most of which can be distilled into: Don't follow fads, the latest shiny tech; they will only break your heart and maybe ruin your career. You want to deliver a quality product that just works, never breaks down, and if it does it is easily diagnosed.
What that means is (a) use software from vendors you can trust and have actual live support (this means they are usually expensive) (b) KISS. In a small shop you will be responsible for everything. You want as few moving parts as possible. You especially want to avoid all custom integrations; they suck. If software B doesn't offer an out of the box connector to software A which your company already is using. They are immediately ruled out. (c) never trust a salesman/saleswoman. Before you sign any contract, you want to try it out as proof of concept. Just because they say they have an out of the box connector, for example, does it actually sync the fields you want? Verify it works before signing any contract. If the salesman doesn't like it, tell them good bye.
Some good advice here, but sometimes writing a connector that pumps data from one REST interface to another one is way simpler than using software B (which has the connector, but is buggy and requires paid upgrades for every integration) instead of software C which has a well documented interface.
KISS actually means, sometimes write it from scratch yourself, which will save you time short-term and teach you great stuff long-term, rather than use the "industry standard".
When setting up my current operation, I needed a job scheduling framework. I looked at Airflow, but by the time I had to choose a DB backend and create usernames and set up SMTP server, I gave up. In half a day I knocked together a basic single machine scheduler, which does 5% of what Airflow does. But it does that very well, took me frankly less time than setting up this behemoth, and works really well for my team.
Is it better than Airflow, or Prefect, or whatnot? Surely not; but it saved me time and taught me a lot about job scheduling.
I didn’t say anything about not writing code yourself. Sometimes that is the best choice. KISS to me means simplicity and less moving parts. If you have to hire a large team to support your data flow, I would argue you are doing KISS wrong.
Most of which can be distilled into: Don't follow fads, the latest shiny tech; they will only break your heart and maybe ruin your career. You want to deliver a quality product that just works, never breaks down, and if it does it is easily diagnosed.
What that means is (a) use software from vendors you can trust and have actual live support (this means they are usually expensive) (b) KISS. In a small shop you will be responsible for everything. You want as few moving parts as possible. You especially want to avoid all custom integrations; they suck. If software B doesn't offer an out of the box connector to software A which your company already is using. They are immediately ruled out. (c) never trust a salesman/saleswoman. Before you sign any contract, you want to try it out as proof of concept. Just because they say they have an out of the box connector, for example, does it actually sync the fields you want? Verify it works before signing any contract. If the salesman doesn't like it, tell them good bye.