Related Posts
I am on bench from 1 nov 2022. I am PA.
Need 11 likes to unlock DM
What is EBA IN Bosch?
Additional Posts in Data & Analytics Consultants
Got messaged by a C3 . ai recruiter. Read that wlb is bad and that the interview process is absurdly long, but the Glassdoor reviews are 4.2 and can't find actual hours worked posted by anyone. How's the culture really? I'd be aiming for DS consulting, something more functional but with DS/ML concepts as my differentiator.
C3.ai, Inc.
New to Fishbowl?
unlock all discussions on Fishbowl.



select * from <data>
Here is your report
Coach
Slalom I be doing that on data sets I’m supposed to know lol. Field names and what not
Planning to do basic counts, uniques, km figure out unique identifiers, and the underlying data model. What else would you suggest?
Selecting / analyzing Distinct values from each column so you know the details of dimensional columns is a must (understanding what makes the categorical data categorical, giving yourself fact data more meaning)
If they can’t hand you a logical data model or a data dictionary then I would build the report with clean dummy data and tell them once you get the data piece figured out we can pour into the dashboard.
Not sure what you’re SoW would read but I hope you didn’t get hung out to dry like thag
No queries. Ask for an ERD diagram & data dictionary first
They don’t have that. What’s next
What idiots.
Bunch of describe queries. Describe database, schema, table.
You’ll get to know the primary keys, data type etc very quickly
*I meant the company who made the database without documentation is an idiot. Not the question or answers in this thread.
All the best!
Figure out what business function it belongs to. Then, based on that, figure out what you are looking at.
Every function always has key data points it uses for kpis.
Sales Orders
Puechase Orders
Invoices
Bills of Landing
Production Orders
SKU
Then, you figure out the right dimensions. Which customer, category, site, account has the top 10 N XXX and how it varies.
This the appropriate approach. Talk to people who may have worked with similar data, look up online for any metrics that is common in the industry, discuss with your team internally to see if they have any inputs, identify hierarchy’s and try to take a first pass at building some of the metrics out. You’ll start seeing patterns and get more ideas as you continue to build metrics out.
Ask for a Entity Relationship Diagram? If they don’t know the data, I’m sure someone designed and normalized the database, right? Come on.
Identify the biggest tables by row count - Customer, Sale, whatever….try to see if there’s logical context you can break down there. Then just do what people have said above.
Look for common identifiers between tables. Could help to identify child records. Customer Table has and ID used in transactions tables.