Arshak Mkhoyan, Developer in Yerevan, Armenia
Arshak is available for hire
Hire Arshak

Arshak Mkhoyan

Verified Expert  in Engineering

Data Scientist and Machine Learning Developer

Yerevan, Armenia

Toptal member since September 6, 2021

Bio

Arshak has 3+ years of experience working as a data scientist and machine learning developer. He helped businesses become more profitable by increasing the click-through rate through recommendation systems and the retention rate of marketing campaigns using uplift modeling. Arshak is looking for projects that allow him to work with data to get valuable insights and develop machine learning algorithms to solve business tasks, showing his proficiency in Python, machine learning, and deep learning.

Portfolio

USMALL
Recommendation Systems, Python, FAISS, Deep Learning, Data Analysis, Pandas...
Globus
Data Science, Machine Learning, SQL, Microsoft Excel, Python, Docker...
BetConstruct
Python, Data Science, Machine Learning, SQL, Deep Learning...

Experience

Availability

Part-time

Preferred Environment

Jira, Slack, Jupyter Notebook, PyCharm, Linux, MacOS

The most amazing...

...project I've developed is a model for predicting customers' uplift (effect of communication) that increased the profitability of the marketing campaign by 20%.

Work Experience

Machine Learning Developer

2021 - 2021
USMALL
  • Developed an item-to-item recommendation system leveraging deep learning methods.
  • Researched eCommerce recommendation systems, state-of-the-art algorithms.
  • Analyzed user traversal patterns, item popularity, and user-item interactions.
  • Used AWS Sagemaker for model testing and training.
Technologies: Recommendation Systems, Python, FAISS, Deep Learning, Data Analysis, Pandas, NumPy, Scikit-learn, Gensim, Data Visualization, Machine Learning, Data Science, Jupyter Notebook, PyCharm, Linux, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Dashboards, Plotly, Docker Hub, Statistical Analysis, Jupyter, Data Analytics, Amazon Web Services (AWS), Data Pipelines, Data Engineering

Senior Data Scientist

2020 - 2021
Globus
  • Developed a model for estimating customers' uplift in marketing campaigns, increasing their profitability by 20%.
  • Created a churn prediction model which assists company operators in predicting customers who are most likely subject to churn.
  • Built a tool for finding similar groups of users to the one requested based on specific features.
  • Analyzed customer behavior, goods characteristics, and hypermarkets.
Technologies: Data Science, Machine Learning, SQL, Microsoft Excel, Python, Docker, Data Analysis, Statistics, Pandas, NumPy, Scikit-learn, Data Visualization, Jupyter Notebook, PyCharm, Dashboards, Plotly, Flask, Docker Hub, Statistical Analysis, Jupyter, Data Analytics, Data Scraping, Data Pipelines, Data Engineering

Machine Learning Developer

2019 - 2021
BetConstruct
  • Created an extractive question answering system based on NLP and statistical methods.
  • Built a bot detection system to identify parsers among regular users of the website.
  • Developed a chatbot system integrating several APIs and NLP-based methods.
  • Managed a team of three data scientists using Jira as a reporting tool.
  • Created a poker AI after researching reinforcement learning and game theory.
Technologies: Python, Data Science, Machine Learning, SQL, Deep Learning, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Amazon Web Services (AWS), Docker, Git, FAISS, Elasticsearch, Google Cloud Platform (GCP), MongoDB, Selenium, PyTorch, Pandas, NumPy, Scikit-learn, Gensim, Data Visualization, Data Analysis, TensorFlow, Jira, Slack, Jupyter Notebook, PyCharm, Linux, Dashboards, Plotly, Flask, Statistical Analysis, Jupyter, Data Analytics, Statistical Modeling, Data Scraping, Data Pipelines, Data Engineering

Business Analyst

2018 - 2018
Ameriabank CJSC
  • Conducted weight adjustment on multiple survey data, ensuring accurately representative samples.
  • Analyzed available data by identifying main trends, changes, and root causes.
  • Visualized data and prepared reports for stakeholders.
Technologies: Python, Excel 365, Tableau, HTML Parsing, Pandas, Matplotlib, Dashboards, Jupyter, Data Analytics, Business Intelligence (BI), Statistical Modeling, Data Scraping

Uplift Model for Marketing Campaigns

Globus needed accurate information about the effect of their marketing campaigns' communication on customers' behavior.

I developed a machine learning model for predicting customers' uplift in marketing campaigns. To built the model, I collected the data via A/B testing, extracted the insights, chose ML and business metrics to track performance, prepared data for modeling, created the uplift model, and tested this model in offline and online environments.

Overall, this model helped to increase the profitability of marketing campaigns by 20%.
2014 - 2019

Bachelor's Degree in Economics

American University of Armenia - Yerevan, Armenia

OCTOBER 2021 - PRESENT

Google Cloud Big Data and Machine Learning Fundamentals

Google Cloud | via Coursera

JULY 2019 - PRESENT

Introduction to Deep Learning

HSE University | via Coursera

MARCH 2019 - PRESENT

Machine Learning

Stanford University | via Coursera

Libraries/APIs

Pandas, NumPy, Scikit-learn, TensorFlow, PyTorch, Matplotlib

Tools

Jira, Slack, PyCharm, Microsoft Excel, Git, Gensim, Jupyter, Plotly, Docker Hub, Tableau, BigQuery, Cloud Dataflow

Languages

Python, SQL

Frameworks

Selenium, Flask, Hadoop

Platforms

Jupyter Notebook, Docker, Linux, MacOS, Google Cloud Platform (GCP), Amazon Web Services (AWS)

Paradigms

Business Intelligence (BI)

Storage

MongoDB, Data Pipelines, Elasticsearch

Other

Data Analysis, Data Science, Deep Learning, Machine Learning, Data Visualization, Statistical Analysis, Data Analytics, Statistical Modeling, Data Scraping, Economics, Natural Language Processing (NLP), Recommendation Systems, FAISS, Statistics, Data Engineering, Generative Pre-trained Transformers (GPT), A/B Testing, Excel 365, HTML Parsing, Dashboards, Big Data

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring