Posts

Creating a good looking dashboard: 1. Sketch out overall design 2. Futura font 3. Limit to few colors 4. Always have a general card system 5. Chart should have meaning 6. Create questions 7. Key metrics are:     + Revenue     + Profit     + By time, by method, distributor     + Answer questions  
 https://www.youtube.com/watch?v=pMQSgriJqR8  Lore So What: How much SQL, PowerBI Tableau, Python do you actually need in Data Analyst Job SQL - ESSENTIAL   1. Currently, in my opinion it is PREFERABLE to use PostgreSQL:   ✔ From Reddit reviews it is faster, cleaner that MySQL, SSMS   ✔ Oracle is not really mentioned   ✔ UI is better looking   ✔ GitLab using PSQL 2. Comfortable using tools:   + Add a lot of comments to run the SQL Query independently by selecting only the needed part   + Basic concepts + Syntax + Basic commands + Aggregate manipulation + Do constant Sorting   + CTEs - very important and also simple. It basically creates a temporary table with all values needed   + Windows function - Generally the Aggregates + OVER() (PARTITION BY column) - extremely useful   + Null Values with Coalesce   + Joining tables    + String Manipulation: CONCAT, LOWER, UPPER, DATEDIFF, DATEADD, EXTRACT, YEAR, MONTH, ROUN...
 https://www.youtube.com/watch?v=NtfbWkxJTHw Fireship: 1. In coding it is much more important to learn pattern and tools for problem solving skills 2. Feyman technique: teaching your own 12 years old self on the process and method  (self mentoring) 3. Refactor and clean, so others could teach you