Preguntas de entrevista de Gerente de ciencia de datos

246

Preguntas de entrevista para Gerente De Ciencia De Datos compartidas por los candidatos

Principales preguntas de entrevista

Ordenar: Relevancia|Popular|Fecha
Meta
A un Data Science Manager le preguntaron...7 de octubre de 2020

A table with posts and a table with reactions to posts. Find the avg number of likes per post for the past 28 days.

4 respuestas

You have a metric, then some UI have changed over time, your metric increased by 20%. How do you measure the change?? This was what he said word for word. Indeed it sounded weird to me at first: the changed was measured, its 20%, but I went ahead and assumed the question was on attribution analysis. But eventually it seems the question was just how to do a AB test if something needs to change. Menos

Yeah, odd. They probably wanted something like a pre-post analysis I guess with a t-test to show significance in the difference? But that should never be the case - you should always experiment first you said. Menos

Select p.postid, ,sum(case when reaction =’likes’ then 1 else 0 end) /count(*) Posts p Join likes r on p.postid = r.postid where data = currentdate – interval ’28 day’ group by p.postid, Menos

Mostrar más respuestas
DISH

Do you have experience managing projects involving developers located on another continent?

2 respuestas

yes, when working with off-shore teams it's critical to ensure that the problem is clearly defined, don't leave anything to guess work. Agile works best for off-shore to ensure resource/team is making good progress, keep on task and can provide guidance/clarification when needed early on. Also recommend that there is an off-shore project manager that you work with if possible. Menos

yes, when working with off-shore teams it's critical to ensure that the problem is clearly defined, don't leave anything to guess work. Agile works best for off-shore to ensure resource/team is making good progress, keep on task and can provide guidance/clarification when needed early on. Also recommend that there is an off-shore project manager that you work with if possible. Menos

Symphony Teleca

Case study on Retail analysis case. As I was from BFSI Domain this was new to me, but interesting so.

1 respuestas

From statistical and basic knowledge on online shopping I could at least try possible solutions for the question posed. I think that was very close to what they expected. Menos

University of Utah

Do you have experience as a manager?

1 respuestas

I had experience as a manager and explained my relevant experience

IRI

Explain one of my projects

1 respuestas

Briefed my project.

naviHealth

A working problem to see how I would approach a data science project including project management, MVP, data wrangling, EDA, model choices, and interpretation goals.

1 respuestas

Project Management for Data Science can vary depending on your project, but the DS standard project Phases are outlined below: Data Analytics/Data Science project phases • Business Understanding or Problem definition • Data Interpretation • Data Preparation • Modeling • Evaluation • Deployment Minimum viable product (MVP) – is a version of a product with just enough features to satisfy early customers adoption and provide feedback for future product enhancements or development efforts. Gathering insights from an MVP is often less expensive than developing a product with more features, which increases costs and risk if the product fails, for example, due to incorrect assumptions. It may also involve carrying out market analysis beforehand. When using Agile Following these steps will help you identify and prioritize features, as well as help you confidently outline what you need to get your MVP to market. Step 1: Identify and Understand your Business and Market Needs, goals and objectives/KPIs Step 2: Map Out User Journey(s) ... Step 3: Create a pain and gain map. ... Step 4: Decide What Features to Build. Exploratory data analysis (EDA) – In statistics, exploratory data analysis is an approach to analyzing data sets to summarize their main characteristics, often with visual methods. A statistical model can be used or not, but primarily EDA is for seeing what the data can tell us beyond the formal modeling or hypothesis testing task. Exploratory data analysis was promoted by John Tukey by to encourage statisticians to explore the data, and possibly formulate hypotheses that could lead to new data collection and experiments. EDA is different from initial data analytis (IDA) which focuses more narrowly on checking assumptions required for model fitting and hypothesis testing, and handling missing values and making transformations of variables as needed. EDA encompasses IDA. Data Wrangling (Data Preparation Steps) • Raw Data – Define, Locate, Extract • Structure Data – Data transformation, Cleaning and pull into a cube, database, excel spreadsheet or some other storage format. • Data Preprocessing - Data aggregation, data preprocessing prepares data for EDA. • Exploration Data Analysis (EDA) – Modeling, training data • Insight, Reports, Dashboards, Visual Graphics – pulling data into a visual Menos

Pagaya

Statistics: X, Y ~ N(0, 1), What is P(X+Y>=0.5).

1 respuestas

2. Two pretty standard dynamic programming questions.

Flatiron Health

The technical interview required live coding. I don't recall all the details, some combinatorial problem that probably should've been easy but it's been a while since I've coded and was a bit caught off-guard by having to write code when interviewing for this job, which by all accounts would not involve individual technical contribution.

1 respuestas

Very very slowly.

Cognizant Technology Solutions

How do you stay up to date with the markets?

1 respuestas

through online courses in Udemy / coursera

Komodo Health

They asked a Bayesian question.

1 respuestas

Bayesian statistics is a mathematical approach to calculating probability in which conclusions are subjective and updated as additional. data is collected. Bayesian data scientists first analyze sample data and draw a conclusion. This is called the prior inference. Menos

Viendo 1-10 de 246 preguntas de entrevista