Respect Failed People - Knowledge also comes with Failure

Failed people/failed entrepreneurs will have more lessons than accomplished people. Not earning much, Not well studied / not being so successful does not mean the person has limited thinking or expertise. Passion is poisonous. 

What Failed people can teach?

• Where you have to be careful
• Who could potentially be your silent competitor
• What things look similar to his execution
• How time played a role
• How tech played a role
• What you mistrust
• Where you seek more support
• What thinks they overlooked

Look beyond normal perspectives.

Question to understand product mindset / Experience

• How many times you have to had thought to build MVP that beats current competitors/products you felt could be better
• What key lessons have you learned with bad UX, Any experiences?
• What product would you rebuild from your past projects, if you have to go back 10 years and use the latest tech tools?
• Code vs Architect vs Learn domain - What is important with experience?
• What lessons you picked from a major failed project 

Keep Exploring!!! Curiosity never fails!!!

My Perspective of Interviews

Few things I keep a tab on from a time/candidate perspective - Listening to candidate answers, asking for quantifiable data, cover all areas during the time.

• Project discussions - To bring the best out of candidates I ask them to pick their best projects to demonstrate architecture challenges, performance issues, deployment.
• Introduce certain scenarios/brainstorm to get perspectives from candidates. I look at areas they are able to explore / with constraints the alternatives they bring to the table.

Make it a good experience for the candidate, We all keep learning. Be better than yesterday.  

Keep Thinking!!!

Lecture Notes - Google Machine Learning System Design Mock Interview

 

Key Notes

• Billions of videos, Millions of users
• Recommend video for right user / time / context
• Start small, Scale to current needs
• Recommendation based on the entire corpus of content
• Meat in the middle approach
• Pick a Small dataset
• Video, User1, Completion %%
• Video, User2, Completion %%
•  Use for a similar taste
• Similarity-based on some model
• The user doesn't like video
• Collaborative filtering techniques
• Activity, Attributes, Demographics
• User - User
• Demographics, Age, locale, Historical data
• Weightage of recent vs previous topics
• Topic Continuity based recommendations
• Reference / Approximation
• Comparing across videos/users
• Auto recommendations / Next list to display
• Map user/item vector and compare with Cartesian steps
• Edge cases - Cold start / new user content
• The initial filter of videos based on preferences then include trending videos in the domain, Add scoring
• Use choices long/short videos 
• Ranking step / Filtering Step

My thoughts

• Clustering and they recommending based on cluster they belong
• Custom models for each locale, interests then getting it prioritized

Machine Learning System Design (YouTube Recommendation System)

Key Notes

Multitask ranking system

• Two stage pattern
• Select narrowed down candidates
• Ranking approach to check those candidates
• Funnel Fashion
• Could be SQL query / Watched by users
• Regression to narrow down further
• Finding relevant candidates for users

• Current watched video + context
• 700 users / second / scalable system

• Predict the probability of engagement for a video
• For each value multiple predictions
• Combine all outputs, ensemble them

Keep Thinking!!!