职位描述

About Kayzen

Kayzen is a mobile demand-side platform (DSP) that aims to democratize programmatic advertising by enabling apps, agencies, media buyers, and brands to manage programmatic campaigns for customer acquisition, retargeting, and brand performance. The platform is built on performance, transparency, and control, handling over 160 billion daily ad requests from 1.6 billion users globally, serving more than 1 billion ads every day across 180 countries. It offers both API and user interface access.

Team and Role

You will join a diverse and globally distributed Data Science team focused on deploying machine learning models that analyze terabytes of data and serve billions of ad requests and users in real time. This role involves contributing to innovative data science solutions in the fast-evolving ad-tech industry.

Responsibilities

Continuously enhance existing machine learning models by integrating new features, fine-tuning parameters, and leveraging supplementary data.
Collaborate with Machine Learning Engineers to develop and implement scalable supervised and unsupervised learning algorithms.
Explore and identify new data sources, researching innovative approaches to utilize both new and existing datasets effectively.
Investigate and propose novel machine learning solutions aimed at optimizing processes across the value chain.
Conduct and interpret A/B tests regularly to extract insights and improve strategies.

Data Science Principles

Prioritize simple, effective solutions tailored to specific problems rather than forcing data to fit pre-existing models.
Focus on impactful machine learning applications that drive measurable product improvement and client success.
Value collaborative work, requiring strong communication within a multidisciplinary team including analysts, engineers, and scientists.
Embrace experimentation as a core approach to innovation and continuous improvement.

Requirements

A minimum of three years of relevant experience building data science products from concept through production deployment.
Advanced coding skills emphasizing clean, reproducible, and thoroughly tested codebases.
Proficiency with Python, Spark, Hadoop, Airflow, Docker, and SQL.
Experience working with large, sparse datasets using algorithms for prediction, clustering, and outlier detection.
Knowledge of Gradient Boosting Trees frameworks like CatBoost and LightGBM in production settings is advantageous.
Hands-on experience developing neural network-based products for classification, regression, and multi-task learning is highly desirable.
Additional expertise in reinforcement learning and large-scale optimization problems is a plus.
Strong ability to create and improve dashboards and monitoring tools using SQL.
Prior background in the Ad-Tech industry is mandatory.

Benefits

Direct collaboration with company founders, providing a significant chance to influence outcomes.
Outstanding opportunities for professional growth and skills enhancement.
Work alongside a global team of seasoned industry experts and entrepreneurs driving transformation in the Ad-Tech space.
Join a vibrant, international workforce distributed across Germany, India, Argentina, Canada, Spain, the UK, and other countries.
Flexible remote work arrangements with an option to relocate to the Berlin office.
A $500 budget for home office setup.
An annual $1000 learning and development allowance.

Data Scientist