Principal Machine Learning Engineer, TEAM
DoorDash · New York, NY · 1 mo ago
Engineering$282k–$415k/yrFull-time
We need to output HTML fragment with sections: About the team? Actually we have "About The Team" and "About The Role". The instruction: use for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). Do NOT repeat the job title as a heading. We should include sections only when posting has real specific content.
We have content: About The Team (description of team). About The Role (detailed responsibilities). Then sections like "Example Focus Areas Include" (list of focus areas). Then "We’re Excited About You Because You Have…" (qualifications). Then "Compensation" (pay range). Then benefits description. Then other sections like "About DoorDash", "Our Commitment to Diversity and Inclusion", etc. Those are boilerplate? Might be considered generic company info; but we need to preserve substantive information, remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. The "About DoorDash" and diversity statement may be considered boilerplate? The instruction says remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate. The diversity statement is EEO; likely remove. The "About DoorDash" maybe generic but could be considered substantive? It's about mission; maybe keep? The instruction: "Prefer a few substantive sections over many thin ones." We can include About The Team, About The Role, Responsibilities (maybe part of About The Role), Requirements/Qualifications (We’re Excited About You...), Benefits, Pay. Also maybe "Example Focus Areas Include" could be part of Responsibilities or a separate section like "Focus Areas". Could be a subsection under Responsibilities.
We need to output only HTML fragment, no markdown fences.
We'll structure:
for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). Do NOT repeat the job title as a heading. We should include sections only when posting has real specific content.
We have content: About The Team (description of team). About The Role (detailed responsibilities). Then sections like "Example Focus Areas Include" (list of focus areas). Then "We’re Excited About You Because You Have…" (qualifications). Then "Compensation" (pay range). Then benefits description. Then other sections like "About DoorDash", "Our Commitment to Diversity and Inclusion", etc. Those are boilerplate? Might be considered generic company info; but we need to preserve substantive information, remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. The "About DoorDash" and diversity statement may be considered boilerplate? The instruction says remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate. The diversity statement is EEO; likely remove. The "About DoorDash" maybe generic but could be considered substantive? It's about mission; maybe keep? The instruction: "Prefer a few substantive sections over many thin ones." We can include About The Team, About The Role, Responsibilities (maybe part of About The Role), Requirements/Qualifications (We’re Excited About You...), Benefits, Pay. Also maybe "Example Focus Areas Include" could be part of Responsibilities or a separate section like "Focus Areas". Could be a subsection under Responsibilities.
We need to output only HTML fragment, no markdown fences.
We'll structure:
About the Team
... content ...
About the Role
... content ...
Maybe we split responsibilities into- list.
We have many bullet-like sentences in "You will..." list. We'll turn those into
- .
Similarly "Example Focus Areas Include" list.
Qualifications list: "We’re Excited About You Because You Have…" bullet list.
Compensation: we have pay range and description.
Benefits: description of benefits.
We'll avoid duplication.
Let's extract text.
About The Team:
"DoorDash is building the next generation of causal decisioning systems for New Verticals: grocery, convenience, retail, alcohol, pets, flowers, and other emerging categories. These businesses operate in high-dimensional, dynamic marketplaces where every consumer, merchant, item, promotion, substitution, search result, and delivery promise creates a causal question. We are hiring a Principal Machine Learning Engineer to lead the Causal ML pod and establish the technical foundation for company-level causal decisioning. This is a senior technical leadership role for a practitioner who has built consequential causal systems in production and can turn ambiguous business questions into a coherent measurement and decision platform. A central mandate is to define and build a durable company-level causal value metric: a trusted, long-term signal that estimates the incremental value created by product, growth, and marketplace actions. The metric will connect experiments, observational evidence, and production ML so leaders and product teams can compare investments on a common basis while protecting customer experience and marketplace health."
That's substantive.
About The Role:
"You will set the multi-year technical direction for causal ML, lead the pod’s portfolio and operating model, and remain close to the hardest modeling and systems work. You will be accountable for both scientific credibility and production impact. You’re Excited About This Opportunity Because You Will…"
Then list:
- Lead the Causal ML pod across technical strategy, architecture, execution, and quality.
- Create a roadmap that joins foundational platform work with high-value product applications.
- Define the company-level causal value metric and its measurement framework, including the target construct, time horizon, component outcomes, identification strategy, calibration, uncertainty, and guardrails.
- Build the metric into a decision system that teams can use for product prioritization, experiment readouts, intervention selection, budget allocation, and portfolio tradeoffs.
- Establish how randomized experiments, quasi-experiments, observational estimation, and learned models work together. Make the limits of each source of evidence explicit.
- Architect reusable causal capabilities for treatment effect estimation, surrogate validation, counterfactual policy evaluation, sensitivity analysis, and long-term outcome forecasting.
- Guide production applications across promotions, lifecycle interventions, ranking, recommendations, search, substitutions, demand shaping, and inventory-aware discovery.
- Set standards for validation, monitoring, reproducibility, and governance so causal estimates remain reliable as policies, populations, and marketplace conditions change.
- Influence senior leaders across Product, Engineering, Analytics, Finance, Strategy, and business teams by translating complex causal evidence into clear decisions and tradeoffs.
- Develop senior engineers and scientists through technical direction, design review, coaching, and a high bar for causal reasoning and engineering craft.
Then "Example Focus Areas Include" list:
- Company-level causal value: Create a common, causally grounded measure of the long-term value generated by product and business actions, enabling teams to compare opportunities across surfaces while preserving interpretable components and guardrails;
- Consumer action and lifetime value: Estimate which interventions truly grow durable customer value rather than pull demand forward, including promotions, lifecycle nudges, personalization, retention, and reactivation;
- Surrogate metrics and faster learning: Develop and validate early indicators that accelerate decisions before long-term outcomes mature, paired with variance reduction, sequential learning, and clear standards for trust;
- Counterfactual ranking and decisioning: Build causal layers for recommendations, search, targeting, demand shaping, and marketplace allocation so systems optimize incremental outcomes rather than predictive relevance alone; and
- Causal measurement platform: Create shared infrastructure that connects experiments, observational data, policy logs, estimation, calibration, and decision workflows across teams.
Then "We’re Excited About You Because You Have…" list:
- Extensive experience (10+ years) in causal inference, econometrics, experimentation, or causal machine learning, with a record of setting technical direction beyond a single team.
- Experience leading the design and productionization of causal models, measurement platforms, experimentation systems, or large-scale decision engines.
- Deep judgment about randomized experiments, observational methods, surrogate endpoints, and model-based decisioning, including when evidence is insufficient for a decision.
- Fluency with methods such as doubly robust estimation, double machine learning, instrumental variables, difference-in-differences, synthetic controls, variance reduction, heterogeneous treatment effects, contextual bandits, and off-policy evaluation.
- Strong ML engineering and systems ability. You can shape data contracts, modeling pipelines, evaluation frameworks, serving patterns, and monitoring for high-stakes production use.
- The ability to reason about long-term customer and marketplace value, not only local model metrics or immediate conversion.
- A track record of influencing executives and senior cross-functional partners through clear problem framing, technical judgment, and evidence.
- A multiplier mindset: you create reusable abstractions, improve decision quality across teams, and raise the technical standard of the people around you.
Compensation:
"The successful candidate’s starting pay will fall within the pay range listed below and is determined based on job-related factors including, but not limited to, skills, experience, qualifications, work location, and market conditions. Base salary is localized according to an employee’s work location. Ranges are market-dependent and may be modified in the future. In addition to base salary, the compensation for this role includes opportunities for equity grants. Talk to your recruiter for more information."
Then benefits:
"DoorDash cares about you and your overall well-being. That’s why we offer a comprehensive benefits package to all regular employees, which includes a 401(k) plan with employer matching, 16 weeks of paid parental leave, wellness benefits, commuter benefits match, paid time off and paid sick leave in compliance with applicable laws (e.g. Colorado Healthy Families and Workplaces Act). DoorDash also offers medical, dental, and vision benefits, 11 paid holidays, disability and basic life insurance, family-forming assistance, and a mental health program, among others. To learn more about our benefits, visit our careers page here."
Then Paid Time Off details:
"See Below For Paid Time Off Details For salaried roles: flexible paid time off/vacation, plus 80 hours of paid sick time per year.For hourly roles: vacation accrued at about 1 hour for every 25.97 hours worked (e.g. about 6.7 hours/month if working 40 hours/week; about 3.4 hours/month if working 20 hours/week), and paid sick time accrued at 1 hour for every 30 hours worked (e.g. about 5.8 hours/month if working 40 hours/week; about 2.9 hours/month if working 20 hours/week)."
Then Pay range:
"The national base pay range for this position within the United States, including Illinois and Colorado. $282,100—$414,800 USD"
We should include Pay section with that range.
We should not include application instructions, EEO, diversity, scam warnings, etc.
We'll produce HTML:
About the Team
...
About the Role
You will set the multi-year technical direction for causal ML, lead the pod’s portfolio and operating model, and remain close to the hardest modeling and systems work. You will be accountable for both scientific credibility and production impact. You’re Excited About This Opportunity Because You Will…
- Lead the Causal ML pod across technical strategy, architecture, execution, and quality.
- Create a roadmap that joins foundational platform work with high-value product applications.
- Define the company-level causal value metric and its measurement framework, including the target construct, time horizon, component outcomes, identification strategy, calibration, uncertainty, and guardrails.
- Build the metric into a decision system that teams can use for product prioritization, experiment readouts, intervention selection, budget allocation, and portfolio tradeoffs.
- Establish how randomized experiments, quasi-experiments, observational estimation, and learned models work together. Make the limits of each source of evidence explicit.
- Architect reusable causal capabilities for treatment effect estimation, surrogate validation, counterfactual policy evaluation, sensitivity analysis, and long-term outcome forecasting.
- Guide production applications across promotions, lifecycle interventions, ranking, recommendations, search, substitutions, demand shaping, and inventory-aware discovery.
- Set standards for validation, monitoring, reproducibility, and governance so causal estimates remain reliable as policies, populations, and marketplace conditions change.
- Influence senior leaders across Product, Engineering, Analytics, Finance, Strategy, and business teams by translating complex causal evidence into clear decisions and tradeoffs.
- Develop senior engineers and scientists through technical direction, design review, coaching, and a high bar for causal reasoning and engineering craft.
Focus Areas
with list. However instruction says prefer few substantive sections over many thin ones. Might be okay to include as part of Responsibilities. Qualifications: we can haveQualifications
with list. Benefits:Benefits
with p. Pay:Pay
with p containing range. We might also include Schedule? Not present. We'll not include About The Team as a heading? It's not in the list but we can still useAbout the Team
as it's a section heading. The instruction says usefor section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits,
- .
Similarly "Example Focus Areas Include" list.
Qualifications list: "We’re Excited About You Because You Have…" bullet list.
Compensation: we have pay range and description.
Benefits: description of benefits.
We'll avoid duplication.
Let's extract text.
About The Team:
"DoorDash is building the next generation of causal decisioning systems for New Verticals: grocery, convenience, retail, alcohol, pets, flowers, and other emerging categories. These businesses operate in high-dimensional, dynamic marketplaces where every consumer, merchant, item, promotion, substitution, search result, and delivery promise creates a causal question. We are hiring a Principal Machine Learning Engineer to lead the Causal ML pod and establish the technical foundation for company-level causal decisioning. This is a senior technical leadership role for a practitioner who has built consequential causal systems in production and can turn ambiguous business questions into a coherent measurement and decision platform. A central mandate is to define and build a durable company-level causal value metric: a trusted, long-term signal that estimates the incremental value created by product, growth, and marketplace actions. The metric will connect experiments, observational evidence, and production ML so leaders and product teams can compare investments on a common basis while protecting customer experience and marketplace health."
That's substantive.
About The Role:
"You will set the multi-year technical direction for causal ML, lead the pod’s portfolio and operating model, and remain close to the hardest modeling and systems work. You will be accountable for both scientific credibility and production impact. You’re Excited About This Opportunity Because You Will…"
Then list:
- Lead the Causal ML pod across technical strategy, architecture, execution, and quality.
- Create a roadmap that joins foundational platform work with high-value product applications.
- Define the company-level causal value metric and its measurement framework, including the target construct, time horizon, component outcomes, identification strategy, calibration, uncertainty, and guardrails.
- Build the metric into a decision system that teams can use for product prioritization, experiment readouts, intervention selection, budget allocation, and portfolio tradeoffs.
- Establish how randomized experiments, quasi-experiments, observational estimation, and learned models work together. Make the limits of each source of evidence explicit.
- Architect reusable causal capabilities for treatment effect estimation, surrogate validation, counterfactual policy evaluation, sensitivity analysis, and long-term outcome forecasting.
- Guide production applications across promotions, lifecycle interventions, ranking, recommendations, search, substitutions, demand shaping, and inventory-aware discovery.
- Set standards for validation, monitoring, reproducibility, and governance so causal estimates remain reliable as policies, populations, and marketplace conditions change.
- Influence senior leaders across Product, Engineering, Analytics, Finance, Strategy, and business teams by translating complex causal evidence into clear decisions and tradeoffs.
- Develop senior engineers and scientists through technical direction, design review, coaching, and a high bar for causal reasoning and engineering craft.
Then "Example Focus Areas Include" list:
- Company-level causal value: Create a common, causally grounded measure of the long-term value generated by product and business actions, enabling teams to compare opportunities across surfaces while preserving interpretable components and guardrails;
- Consumer action and lifetime value: Estimate which interventions truly grow durable customer value rather than pull demand forward, including promotions, lifecycle nudges, personalization, retention, and reactivation;
- Surrogate metrics and faster learning: Develop and validate early indicators that accelerate decisions before long-term outcomes mature, paired with variance reduction, sequential learning, and clear standards for trust;
- Counterfactual ranking and decisioning: Build causal layers for recommendations, search, targeting, demand shaping, and marketplace allocation so systems optimize incremental outcomes rather than predictive relevance alone; and
- Causal measurement platform: Create shared infrastructure that connects experiments, observational data, policy logs, estimation, calibration, and decision workflows across teams.
Then "We’re Excited About You Because You Have…" list:
- Extensive experience (10+ years) in causal inference, econometrics, experimentation, or causal machine learning, with a record of setting technical direction beyond a single team.
- Experience leading the design and productionization of causal models, measurement platforms, experimentation systems, or large-scale decision engines.
- Deep judgment about randomized experiments, observational methods, surrogate endpoints, and model-based decisioning, including when evidence is insufficient for a decision.
- Fluency with methods such as doubly robust estimation, double machine learning, instrumental variables, difference-in-differences, synthetic controls, variance reduction, heterogeneous treatment effects, contextual bandits, and off-policy evaluation.
- Strong ML engineering and systems ability. You can shape data contracts, modeling pipelines, evaluation frameworks, serving patterns, and monitoring for high-stakes production use.
- The ability to reason about long-term customer and marketplace value, not only local model metrics or immediate conversion.
- A track record of influencing executives and senior cross-functional partners through clear problem framing, technical judgment, and evidence.
- A multiplier mindset: you create reusable abstractions, improve decision quality across teams, and raise the technical standard of the people around you.
Compensation:
"The successful candidate’s starting pay will fall within the pay range listed below and is determined based on job-related factors including, but not limited to, skills, experience, qualifications, work location, and market conditions. Base salary is localized according to an employee’s work location. Ranges are market-dependent and may be modified in the future. In addition to base salary, the compensation for this role includes opportunities for equity grants. Talk to your recruiter for more information."
Then benefits:
"DoorDash cares about you and your overall well-being. That’s why we offer a comprehensive benefits package to all regular employees, which includes a 401(k) plan with employer matching, 16 weeks of paid parental leave, wellness benefits, commuter benefits match, paid time off and paid sick leave in compliance with applicable laws (e.g. Colorado Healthy Families and Workplaces Act). DoorDash also offers medical, dental, and vision benefits, 11 paid holidays, disability and basic life insurance, family-forming assistance, and a mental health program, among others. To learn more about our benefits, visit our careers page here."
Then Paid Time Off details:
"See Below For Paid Time Off Details For salaried roles: flexible paid time off/vacation, plus 80 hours of paid sick time per year.For hourly roles: vacation accrued at about 1 hour for every 25.97 hours worked (e.g. about 6.7 hours/month if working 40 hours/week; about 3.4 hours/month if working 20 hours/week), and paid sick time accrued at 1 hour for every 30 hours worked (e.g. about 5.8 hours/month if working 40 hours/week; about 2.9 hours/month if working 20 hours/week)."
Then Pay range:
"The national base pay range for this position within the United States, including Illinois and Colorado. $282,100—$414,800 USD"
We should include Pay section with that range.
We should not include application instructions, EEO, diversity, scam warnings, etc.
We'll produce HTML: