Job Description
Hi there!

We are Semrush, a global IT company developing our own product – a platform for digital marketers. New stars are born here, so don’t miss your chance.

This is our role
Data Scientist for those who want to turn ideas into reality using code, algorithms, and maybe a bit of magic

Tasks in the role

- Create models to provide AI insights from traffic data

- Analyze usage and predict&analyze client behavior

- Conducting experiments, storing results, retraining models, quality monitoring, and alerting

- Conducting code reviews


Requirements
Who we are looking for

- Specialized education (mathematics/physics/computer science)

- Deep understanding of machine-learning algorithms and methods

- Knowledge of probability theory, linear algebra, and mathematical analysis

- Ability to write clean Python code

- Experience with databases (SQL DBs, BigQuery)

- Work experience of at least two years in a similar role

- English at a B2 level or higher


Not required, but a plus

- You have experience with explainable AI

- You have experience with MLFlow.

- You have experience with GitLab CI

- You have experience with GCP and Vertex AI

- You have experience in developing ETL processes with Airflow

- You have experience in deploying ML systems with batch/real-time processing and monitoring (Grafana, Prometheus, Alertmanager)

- Reproducible and well-documented research


A bit about the team
You can get to know the team better at one of the interviews, but some brief information about future colleagues will be useful now.

The AI Lab Team is part of the Competitive Intelligence Unit, which develops products for domain traffic and market analytics. The result of our work are: the AI insights for our tools and internal models. We work with a large amount of data and use machine learning in our solutions.
Some projects we want you to be a part of:

- Churn prediction model

- Traffic chart description

- Insights about potential audience


Our stack:

- Google Cloud Platform

- Data Storage: BigQuery, Google Cloud Storage

- Language: Python

- Tools: Airflow, k8s, Docker, FastAPI, XGboost, Pandas, TensorFlow/Torch, MLFlow

- LLM APIs: OpenAI, Google


We will try to create all the right conditions for you to work and rest comfortably

- It’s up to you to decide what work format works best for you. You can #wfo, #wfh, or mix both

- Flexible working day start

- Unlimited PTO

- Hobby benefit

- Breakfast, snacks, and coffee at the office

- Corporate events

- Training, courses, conferences

- Gifts for employees