THE COMPANY
Color Mix Group of Companies is a diversified holding group with interests in digital advertising, e-commerce, and retail media. Our programmatic advertising division serves millions of ad impressions daily across display, video, and connected TV. We are not a traditional agency. We are a technology-driven team that builds algorithms to buy media smarter, faster, and cheaper. We need a Senior Data Scientist who understands real-time bidding, auction dynamics, and predictive modeling.
THE PROBLEM YOU WILL SOLVE
Every millisecond, our bidding systems decide which impressions to buy and how much to bid. The winning bidder gets the impression. The loser moves on. We need to predict click-through rates, conversion value, and optimal bid prices with low latency and high accuracy. You will build and maintain the models that power these decisions.
DUTIES
Algorithm Development (50%)
- Design and implement predictive models for CTR, conversion probability, and lifetime value
- Develop bidding algorithms that optimize toward client ROAS and ROI targets
- Work with streaming data (real-time bid requests, user signals, contextual features)
Production & Scaling (25%)
- Deploy models into production using Python, Spark, and cloud infrastructure (AWS/GCP)
- Monitor model performance, feature drift, and data quality
- Optimize inference latency (sub-100ms requirements)
Experimentation (15%)
- Design and analyze A/B tests for algorithm changes
- Evaluate trade-offs between exploration and exploitation
- Provide recommendations to product and engineering teams
Collaboration (10%)
- Partner with software engineers to integrate models into bidding systems
- Communicate findings to non-technical stakeholders (revenue impact, model improvements)
EXPERIENCE
- 5+ years of industry experience in data science, with at least 2 years focused on programmatic advertising, real-time bidding, or ad tech
- Strong proficiency in Python (pandas, numpy, scikit-learn, XGBoost, LightGBM)
- Experience with time series forecasting, classification, and regression models
- Understanding of auction theory, bid shading, or reinforcement learning for bidding
- Experience with SQL and large-scale datasets (BigQuery, Redshift, Snowflake)
- Familiarity with ML deployment tools (Docker, Kubernetes, MLflow)
- Ability to write clean, production-ready code (not just notebooks)
- Strong communication skills (you can explain a log loss curve to a product manager)
- Master's or PhD in Computer Science, Statistics, Economics, or a related field is strongly preferred
- Legal authorization to work in Canada
WHAT WILL HELP YOU STAND OUT
- Experience with real-time bidding (RTB) protocols (OpenRTB)
- Knowledge of media metrics (CPM, CPC, CPA, ROAS, viewability)
- Experience with DSPs (The Trade Desk, DV360, Amazon DSP) or SSPs
- Familiarity with ad fraud detection or brand safety models
- Published research or conference presentations in ad tech or auction theory
Job Type: Full-time
Pay: $155,342.27-$179,034.73 per year
Benefits:
- Dental care
- Disability insurance
- Employee assistance program
- Extended health care
- Paid time off
- Vision care
Application question(s):
- Are you comfortable working with streaming data and sub-100ms latency requirements?
Work Location: In person
