Seeking Alpha

- Users retention rate analysis
- Users churn analysis and prediction(ML modeling)
- Users interaction analysis and predictions of user's propensity to sign-up(ML modeling)
- User journey analysis (Graph analytics)
- Topic modeling of user reviews (NLP)
- Content recommendation (ML modeling)
- Execute and evaluate A/B tests to improve the user journey
- Explain trends across data sources, potential opportunities for growth or improvement, and data caveats for descriptive, diagnostic, predictive (including forecasting), and prescriptive data projects
- Develop dashboards in Looker and define metrics that inform the success of the Product Team
- Explore large, complex, and loosely defined datasets to create actionable insights

Data & Advanced Analytics Technical Lead

GlobalLogic

- Responsible for leadership of key initiatives within the machine learning team
- Working with various ML tasks building architecture, task decomposition, and estimates
- Pre-sale support and presentation of our expertise for the clients
- Collaborate with the team to build internal PoCs

Data Science Lead CoE

- Lead data mining and collection procedures
- Ensure data quality and integrity
- Build advanced analytic systems and predictive models
- Align data projects with organizational goals

Technical Lead Data & Advanced Analytics

SMART business

Lead Data Scientist

- Research and develop machine learning models
- Work with clients to identify opportunities for leveraging company data to drive business solutions.
- Mine and analyze data from company databases to drive optimization and improvement of product development, marketing techniques, and business strategies.
- Working with creating data and machine learning solution architectures.
- Keep up-to-date with the latest technology trends
- Implement new statistical or other mathematical methodologies as needed for specific models or  analysis
-  Lead and manage a team of data scientists that is innovative, collaborative, and customer-focused.

Creator and mentor of SMART Data Science Academy 2017.
Data Science mentor Microsoft Developer School 2017.
Lecturer on UNIT.AI Boost #1: Сервіси, нейромережі, ML підходи.
Speaker "Predictive Analytics World for Industry 4.0", Munich 2018.
Speaker "Deep Learning World Munich 2019"
Creator and mentor of SMART Data Science Academy 2019.

Senior Data Scientist

Henkel

Project for a Supply chain department
- Create a demand forecast tool (Best forecast accuracy in Henkel Europe)
Data management
-Upload and analysis data about final consumers
Reporting
- Create dashboard in Qlikview
- Create tool for sales analysis in MS Excel;
- Regular reporting;
- New SKU listing planning;
- Ad-hoc analysis;

Data Analyst

Procter & Gamble

Create a financial analysis tool for the sales team;
Create a fraud analysis tool for finance department;
Create promo's budget control tool for the finance department.

Danone

- Create and integrate a new predictive model for demand forecasting;
- Create a new tool for demand forecast (Excel);
- Marketing campaign analysis;
- Regular demand forecasting for one of the Ukraine region.

Planning Specialist

SC adidas Ukraine

Internship
- Closely studying and analyzing past sales data to determine trends in productivity;
- Provides analytical insights to deliver a greater share of promotion and distribution.

Highly skilled Lead Data Scientist/Data and Advanced analyst with 8 years of experience in designing, developing, and implementing data-driven solutions to solve complex business problems. Proficient in machine learning, statistical modeling, and data visualization. Excellent communicator with the ability to lead cross-functional teams and stakeholders. Strong track record of delivering high-quality results and driving business growth through data-driven insights.
Microsoft Certified Solutions Associate - Machine Learning
Microsoft Certified Solutions Expert - Data Management and Analytics

I'm excited to share my article on enhancing A/B testing with Pingouin, a user-friendly Python library for statistical analysis!

In this article, I discuss the challenges of interpreting A/B testing data and how Pingouin can help. Pingouin offers a variety of statistical tests, including t-tests, ANOVA tests, and correlation analysis, that can be used to draw meaningful insights from your A/B testing results.

Check out the article here: https://lnkd.in/dwJVwpXY

#abtesting #pingouin #statistics #dataanalysis

Andrii Shchur

medium.com • 7 min read

Enhancing A/B Testing with Pingouin: A Python Statistical Powerhouse

Microsoft AI news
VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Time.

TL;DR: single portrait photo + speech audio = hyper-realistic talking face video with precise lip-audio sync, lifelike facial behavior, and naturalistic head movements generated in real-time.

VASA is a framework for generating lifelike talking faces of virtual characters with appealing visual affective skills (VAS), given a single static image and a speech audio clip. Our premiere model, VASA-1, is capable of producing lip movements that are exquisitely synchronized with the audio and capturing a large spectrum of facial nuances and natural head motions that contribute to the perception of authenticity and liveliness. The core innovations include a holistic facial dynamics and head movement generation model that works in a face latent space and develops such an expressive and disentangled face latent space using videos. Through extensive experiments, including evaluating new metrics, we show that our method significantly outperforms previous methods along various dimensions comprehensively. Our method delivers high video quality with realistic facial and head dynamics and supports the online generation of 512x512 videos at up to 40 FPS with negligible starting latency. It paves the way for real-time engagements with lifelike avatars that emulate human conversational behaviors.

More details:
https://lnkd.in/dD62dnVm

#AI #Innovation

Tiny but mighty: The Phi-3 small language models with big potential

Today, Microsoft announced the Phi-3 family of open models as the most capable and cost-effective small language models available. Thanks to training innovations developed by Microsoft researchers, Phi-3 models outperform models of the same size and the next size up across various benchmarks that evaluate language, coding, and math capabilities.

#AI #LLM #Microsoft

Taplio

Andrii Shchur's Linkedin Analytics

Andrii Shchur

Andrii Shchur's Best Posts (last 30 days)

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Andrii Shchur's Linkedin Analytics

Andrii Shchur

Andrii Shchur's Best Posts .css-qqfgvy{font-size:var(--chakra-fontSizes-md);}(last 30 days)

Want to drive more opportunities from LinkedIn?

Andrii Shchur's Linkedin Analytics

Andrii Shchur's Best Posts (last 30 days)