The Trade Desk
Staff Applied Scientist
How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Staff Applied Scientist
# Applied Scientist – Marketplace (Staff) [Iconic Advertising Technology at Scale]
Our Work
Our impact touches three critical layers of the modern programmatic marketplace.
1. Marketplace Execution
Build the skeleton of scalable strategy execution
- Control systems: Dynamic ad-budget optimisation via quality signals and real-time KPI targeting
- Supply prediction models: Machine learning to forecast inventory availability for buyer/seller alignment
- Decision logic: Algorithmic frameworks that power programmatic buy-side strategy execution at global scale
2. Marketplace Curation
Turn datapoint noise into actionable audience tailoredness
- Recommendation systems: B2B benchmarks and surrogate metrics to link audience relevance with high-quality supply
- Curation frameworks: Tools that blend affiliate-match scoring with portfolio-level transactional constraints
- Inventory optimisation: Balancing geological diversity, publisher economics, and audience pareto-efficiency
3. Inventory Science
Make ad impact quantifiable, interpretable, and trading-ready
- Measurement frameworks: Experiments & causal attribution to quantify business-impinging KPIs
- Trade-off resolution: Transforming complex decisions into clear metric-driven outcomes and guidelines
- Performance transparency: Attribution of advertiser KPI changes back to inventory decisions
Responsibilities
Key owners must unite rigorous science with engineered practical outcomes:
-
Demand-side dynamic allocation → Build & maintain control systems that allocate budgets across millions of sell-side partners without dark patterns. —— Conditions: reactivity to quality signals, real-time adaptions to performance feedback.
-
Causal estimation at scale → Design experiment frameworks using causal inference to attribute winner-takes-all effects on advertiser KPIs. —— Degrade and operationalise trade-offs into GO/NO-GO heuristics.
Reasons to use Rodeo
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
It searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
-
Audience-aware curation → Sys-admin recommendation systems that automate audience discovery across diverse inventory. —— Minimise latency; prioritise fairness around reach/cost tradeoffs.
-
Predictive modelling & mitigation warnings → Develop time-series forecasting models to flag (auto-repair friendly) shifts in supply/demand.
-
Cross-team translation → Bridge gaps between research and production: ** απ cytoplasm ** (the hard parts) are iterated engineering with design docs → quantum gates → neutral models. —— Partner with product and engineering to first define scalability rationally.
-
Strategic progression → Shape the raison d’être of the team via applied science roadmap: —— Identify candidate business risks → societal allocation effects → proportional rewards.
Who You Are
(Ideal candidategithub★ layouts harnesses "greatness above mediocrity")
We seek an APL-2036 (or IRL equivalent):
-
Hyper-confident at the murky nexus of method and machinery. —— Moves swiftly between statistical modelling → A/B rollerscoasts → ML pipelines. —— Agony over ambiguity is your given: there will be no time for you to second-guess.
-
Degree: Pursued an DPhil or doctorate (or at least: pure <-> applied cross-pours).
-
Experience: 7+ yrs. industry-exucing lightmod autione data science OR 5+ yrs. with PhD littoral output.
-
End-to-end conviction: Thrives at the physical limits of transitioning a math model from research-withinusale → production-ahead-by-factors.
Stimulated functions your tech toolbelt most relishes:
✔ Experimental Design
Get help with your application
Your very own career expert that helps elevate your application to the next level.
✔ Causal Inference & Evaluation Econometrics
✔ Recommendation Systems (ranking, filtering & auction economics)
✔ Large-Scale Data Processing: Spark; EMR / Databricks; TensorFlow/Fernet pipelines for O(Lz≈7) clients.
- Product mindset: You plan señales & bleeding edges with actual velocity.
Musts
- Experiential background: Professional-perfection synergy:
- Data Science — featuring,
- Machine Learning,
- Statistical Methodologists,
- Or Biologically Inspired cognate.
- Ownership continuum: Research → Productionize → A/B → Quotaingx
Methods staging immunity:
- Technical libraries: riding R; Python/Spark/Renact/Fairness etc (anything remote like TensorBoardx(Time crime easier).
- Coding hygiene: Code clarity is moral couples.
Rabbit Holes We Just Plain Like (Nice)
Follow your whimsy!
-
Programmatic auction ecosystems & sold-side RTB.
-
Adaptive Real-Time Control: Might relate rightfully to dynamic budget & ‘demand pursuit’ heuristics.
-
‘Explainable AI for Advertisers’: Agent (LLM-free, RL/Pact inspired) engines cementing trust and signal timing.
Must-Disclaimours
-
Recruits shut out avoid those bill-setters left-handed. ✋
-
As our namesake captures, we *build one thing harder:
“The Trade Desk works when the world stops working.”
EΩE Statement & Legalese
Life Within
- Agency (ascension); Embedding; Robust working circumstances.
- GBPO$ (Gender, Belief, Plus Orientation soda) .
- Accessibility-conscious i10 & apprt manual.
- Workforce center first. Terminate at cause.
- Accommodations table: if either’s got needs (email [☁]accommodations@thetradedesk.com).
Cash-pay: Only via direct investigation roles.
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London
Skills
Location