Data Science Intern at DoubleVerify

About the company:

"DoubleVerify Scibids" is a global leader in AI-powered digital campaign activation. Scibids empowers global brands to drive specific KPIs and tangible outcomes more effectively while improving operational efficiency and reducing manual lift. DoubleVerify technology does not rely on digital identifiers such as cookies and can be activated across leading Demand-Side Platforms such as The Trade Desk, DV360, and Xandr to mention a few. To learn more visit https://doubleverify.com/scibids-ai/

Recently, DoubleVerify closed its acquisition of Scibids providing an End-to-end, comprehensive measurement and granular optimization. DoubleVerify is the recognized market innovator with the technology and tools that accurately authenticate the quality of digital media and drive ad performance for the world's largest brands. DV provides media transparency and accountability to deliver the highest level of impression quality for maximum advertising performance.  Since 2008, DV has helped hundreds of Fortune 500 companies gain the most from their media spend by delivering best-in-class solutions across the digital ecosystem that help build a better industry. Learn more at doubleverify.com

"Scibids DoubleVerify" Technology equips these trading desks with an artificial intelligence software computing the optimal amount to bid for each advertising opportunity, thus bringing the final advertisers dramatic performance and productivity uplifts. The technology developed by Scibids is rooted in a solid scientific and market expertise, with one of the founders holding a PhD. in Machine Learning from ENS Cachan and 10 engineers cumulating 40 years of experience in quantitative advertising.  

Our company was selected in 2017 as one of the 62 most innovative startups (amongst 2000+) in the extremely selective “I-Lab” competition by the French public bank & French research ministry. It was also selected in the "Top 100 French Startup To Follow in 2017" by Business Insider.

Two years only after its creation, the company already has several dozens of clients throughout the world and its portfolio keeps on growing! In September 2018, Scibids announced a €2,2M funding in order to support its global ambition to remain the largest algorithmic media trading platform in the world.

What you will do & what we offer

The mission of Scibids is to provide media buyers with fully automated buying strategies for display campaigns, leading to uncompared  performances for a minimum of human work. 

As a DataScientist, your role will be to evaluate and improve "Scibids DoubleVerify" bidding algorithms, which are Scibids main asset, and to actively participate into Scibids automation and development process: each business case not yet automatically handled by our product should be met by the corresponding product improvement. This position will combine analytical rigor, a relentless curiosity, a strong knowledge of statistical methods and experience in implementing them.

This position will combine high development standards, a relentless curiosity, a strong understanding of business priorities and the ambition for your work to become a real game changer!

On a day to day basis, your responsibilities will include:

• working with massive datasets (billions of ads impressions/month) and millions of variables. 
• solving non-routine analysis problems (not handled by off-the-shelf ML solvers).

solving challenging scaling/engineering problems as our system produces thousands of ML models per day.

• working closely with the commercial and customer success team and help identify the business cases where a new algorithm is needed/existing ones need improvement.
• identify and automate any repetitive process taking place at Scibids, including cases where this means developing some AI routines replacing “human common sense”.

Qualifications

Minimum qualifications:

• MS degree in applied mathematics (e.g., statistics, probabilities, operations research, machine learning).
• Work / academic projects experience in data analysis or related fields (e.g., as a statistician/data scientist) with actual algorithms implementation in an industry programming language (e.g. Python, C++...) 

Technical stack relevant to the position

• Python
• SQL
• Kubernetes, Docker, Github

• Google Cloud Environment, App Engine Standard, Cloud SQL, Google Cloud Datastore...

Cultural fit

• Do things only once. No repetitive task survives ingenuity and automation: at Scibids, everything needs to scale!
• Base your decisions on data points. Preconceived personal convictions rarely survive thorough confrontation with actual data.
• Work in a team and be curious. Every Scibids member has stock-options and is focused on the same objective: make Scibids the next adtech unicorn! This means being interested in how one’s daily personal work fits in Scibids general strategy and having in mind the greater good of the company rather than individual accomplishments.
• Have a good laugh. We are a young, hard-working team with huge ambitions: nothing wipes off long hours of work better than sharing a drink with the team!

Others

• Contract type: End of study internship (French “Stage de fin d’étude”).
• Based in Paris (75008) – Boulevard Haussman
• Compensation: 2500 euros / month.