Director of Engineering, Machine Learning Recommendations
Etsy · Brooklyn, NY · 1 mo ago
Information TechnologyFull-time
We need to output HTML fragment with sections: About the role? Actually the posting includes Company Description, What's the role?, What's this team like?, What does the day-to-day look like?, Qualities That Will Help You Thrive, Compensation & Benefits, Location, What's Next, Our Promise etc. We need to use for section headings like About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. However we must only include sections when posting has real specific content. We need to map content to these headings. The posting includes responsibilities (day-to-day), requirements (qualities), qualifications (maybe Qualities That Will Help You Thrive), skills (maybe not explicit), benefits (Compensation & Benefits), pay (Salary Range), schedule (Location mentions in-office days). Also About the role could be the description of role (What's the role?).
We must not repeat job title as heading. No Job ID. Remove boilerplate (EEO, etc). Also remove navigation, cookie, legal, etc. Also remove application instructions (What's Next maybe considered application instruction? It says "If you're interested... share your resume". That's application instruction; we should remove it. Also "Our Promise" is EEO boilerplate; remove.
We need to preserve all substantive info. So we need to include sections: About the role (from What's the role? and maybe team description?), Responsibilities (day-to-day), Requirements (Qualities That Will Help You Thrive), Qualifications (maybe also from Qualities? but we can combine? The instruction says use headings for those categories; we can have multiple sections if content exists. We could have About the role, Responsibilities, Requirements, Qualifications, Benefits, Pay, Schedule. Ensure we don't create sections with filler like "not specified". We have actual content for each.
Let's extract:
About the role: "We are looking for a Director of Engineering, Machine Learning to lead Etsy's Recommendations, the team responsible for the algorithms and systems that power personalized onsite-discovery across our marketplace. This is a pivotal technical leadership role at the intersection of ML research, systems engineering, and product impact. In this role, you will own the end-to-end technical strategy for how Etsy connects buyers with the right listings across the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. You will lead a team of ML engineers, applied scientists, and system engineers organized across three areas: candidate retrieval, ranking, and recommendation systems. You will drive the evolution of our models from state-of-the-art multi-interest retrieval and multi-task deep ranking to next-generation LLM-powered foresight recommendations and generative discovery experiences. The biggest unsolved challenge in front of this team is not just relevance — it is inspiration: helping buyers discover new ideas and start new shopping missions. This opportunity is a full-time position reporting to the VP of Engineering at Search, Recommendations, and Ads."
Also maybe include team description: "The Recommendations organization sits at the heart of Etsy's marketplace. We own over 60 recommendation modules across Etsy's surfaces, with our most impactful work powering the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. Our ML stack spans the full recommendation pipeline..." That could be part of About the role or maybe separate but we can include in About the role.
Responsibilities (day-to-day): list bullet points from "What does the day-to-day look like?":
- Define and drive the multi-year product and technical roadmap for Recommendations ML, with particular focus on personalization, foresight, diversity, and content freshness — the four problems where we have the most room to grow.
- Lead and grow a team of ML engineers, applied scientists, and platform engineers across retrieval, ranking, and systems sub-teams, each with its own engineering manager.
- Partner closely with product, UX, and cross-functional ML teams (Search, Ads, Buyer Understanding, Content Understanding) to align on a coherent, end-to-end buyer experience and shared infrastructure investments.
- Drive the architectural evolution from our current two-stage retrieval-and-ranking pipeline toward more generative and LLM-integrated recommendation architectures.
- Set the technical bar for model development, experimentation, evaluation, and productionization — including our offline VQA and LLM-as-judge evaluation frameworks.
- Champion a high-velocity experimentation culture: our Multi-Variant Dataset (MVD) framework allows parallel A/B tests across retrieval and ranking components, and you should know how to squeeze value out of every experiment.
- Mentor and develop engineering talent at all levels, including Staff and Senior Staff engineers and the engineering managers who report to you.
- Of course, this is just a sample of the kinds of work this role will require! You should assume that your role will encompass other tasks, too, and that your job duties and responsibilities may change from time to time at Etsy's discretion, or otherwise applicable with local law.
We can keep that last sentence as part of responsibilities.
Requirements (Qualities That Will Help You Thrive): list bullet points:
- Proven ability to lead and develop multi-disciplinary teams of ML engineers, applied scientists, and platform engineers, including managing managers and senior managers.
- 10+ years of progressive experience in ML engineering, with demonstrated success shipping large-scale recommendation, retrieval, or ranking systems into production.
- A track record of driving complex, cross-functional technical programs to completion — including navigating shared infrastructure decisions with partner teams in Search, Ads, or adjacent ML platforms.
- Sharp instincts for balancing near-term product impact with mid-to-long-term research investment, and the ability to communicate that trade-off clearly to both engineering and product leadership.
- Familiarity with large-scale ML infrastructure: distributed training (TensorFlow, PyTorch), feature stores, model serving, and online experimentation frameworks.
Qualifications: maybe we could treat the above as Qualifications, but we already used Requirements. Could also have Qualifications separate if there are any explicit qualifications like education? Not present. Could skip Qualifications section if no specific content. But instruction: "Prefer a few substantive sections over many thin ones." We could combine Requirements and Qualifications into one section? But headings are prescribed; we can still include Qualifications if we have content. Since we don't have separate qualifications, maybe we omit Qualifications heading.
Skills: maybe we could extract from Qualities: familiarity with large-scale ML infrastructure etc. Could be Skills. But we already used that in Requirements. Might be redundant. Better to have sections: About the role, Responsibilities, Requirements, Benefits, Pay, Schedule.
Benefits: from Compensation & Benefits: "In addition to salary, you'll be eligible for an equity package, an annual performance bonus, and our competitive benefits that support you and your family." Also maybe mention geographic pay zones? That's Pay.
Pay: Salary Range Geo-1: 296,000.00 - 400,000.00 USD Annual; Geo-2: 266,000.00 - 360,000.00 USD Annual.
Schedule: Location mentions: "We prioritize candidates based within commutable distance of Etsy’s Brooklyn Office. Depending on proximity, we ask team members to come in once or twice per week. Remote candidates outside commutable distance may be considered on a case-by-case basis, with the expectation of occasional travel to the office." That's schedule info (hybrid). So we can put that under Schedule.
We must not include boilerplate like EEO, etc. Also remove "What's Next" and "Our Promise". Also remove "Learn more details about our work modes and workplace safety policies here." (that's a link). Also remove "While Etsy supports visa sponsorship..." etc.
Now produce HTML fragment with
for section headings like About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. However we must only include sections when posting has real specific content. We need to map content to these headings. The posting includes responsibilities (day-to-day), requirements (qualities), qualifications (maybe Qualities That Will Help You Thrive), skills (maybe not explicit), benefits (Compensation & Benefits), pay (Salary Range), schedule (Location mentions in-office days). Also About the role could be the description of role (What's the role?).
We must not repeat job title as heading. No Job ID. Remove boilerplate (EEO, etc). Also remove navigation, cookie, legal, etc. Also remove application instructions (What's Next maybe considered application instruction? It says "If you're interested... share your resume". That's application instruction; we should remove it. Also "Our Promise" is EEO boilerplate; remove.
We need to preserve all substantive info. So we need to include sections: About the role (from What's the role? and maybe team description?), Responsibilities (day-to-day), Requirements (Qualities That Will Help You Thrive), Qualifications (maybe also from Qualities? but we can combine? The instruction says use headings for those categories; we can have multiple sections if content exists. We could have About the role, Responsibilities, Requirements, Qualifications, Benefits, Pay, Schedule. Ensure we don't create sections with filler like "not specified". We have actual content for each.
Let's extract:
About the role: "We are looking for a Director of Engineering, Machine Learning to lead Etsy's Recommendations, the team responsible for the algorithms and systems that power personalized onsite-discovery across our marketplace. This is a pivotal technical leadership role at the intersection of ML research, systems engineering, and product impact. In this role, you will own the end-to-end technical strategy for how Etsy connects buyers with the right listings across the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. You will lead a team of ML engineers, applied scientists, and system engineers organized across three areas: candidate retrieval, ranking, and recommendation systems. You will drive the evolution of our models from state-of-the-art multi-interest retrieval and multi-task deep ranking to next-generation LLM-powered foresight recommendations and generative discovery experiences. The biggest unsolved challenge in front of this team is not just relevance — it is inspiration: helping buyers discover new ideas and start new shopping missions. This opportunity is a full-time position reporting to the VP of Engineering at Search, Recommendations, and Ads."
Also maybe include team description: "The Recommendations organization sits at the heart of Etsy's marketplace. We own over 60 recommendation modules across Etsy's surfaces, with our most impactful work powering the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. Our ML stack spans the full recommendation pipeline..." That could be part of About the role or maybe separate but we can include in About the role.
Responsibilities (day-to-day): list bullet points from "What does the day-to-day look like?":
- Define and drive the multi-year product and technical roadmap for Recommendations ML, with particular focus on personalization, foresight, diversity, and content freshness — the four problems where we have the most room to grow.
- Lead and grow a team of ML engineers, applied scientists, and platform engineers across retrieval, ranking, and systems sub-teams, each with its own engineering manager.
- Partner closely with product, UX, and cross-functional ML teams (Search, Ads, Buyer Understanding, Content Understanding) to align on a coherent, end-to-end buyer experience and shared infrastructure investments.
- Drive the architectural evolution from our current two-stage retrieval-and-ranking pipeline toward more generative and LLM-integrated recommendation architectures.
- Set the technical bar for model development, experimentation, evaluation, and productionization — including our offline VQA and LLM-as-judge evaluation frameworks.
- Champion a high-velocity experimentation culture: our Multi-Variant Dataset (MVD) framework allows parallel A/B tests across retrieval and ranking components, and you should know how to squeeze value out of every experiment.
- Mentor and develop engineering talent at all levels, including Staff and Senior Staff engineers and the engineering managers who report to you.
- Of course, this is just a sample of the kinds of work this role will require! You should assume that your role will encompass other tasks, too, and that your job duties and responsibilities may change from time to time at Etsy's discretion, or otherwise applicable with local law.
We can keep that last sentence as part of responsibilities.
Requirements (Qualities That Will Help You Thrive): list bullet points:
- Proven ability to lead and develop multi-disciplinary teams of ML engineers, applied scientists, and platform engineers, including managing managers and senior managers.
- 10+ years of progressive experience in ML engineering, with demonstrated success shipping large-scale recommendation, retrieval, or ranking systems into production.
- A track record of driving complex, cross-functional technical programs to completion — including navigating shared infrastructure decisions with partner teams in Search, Ads, or adjacent ML platforms.
- Sharp instincts for balancing near-term product impact with mid-to-long-term research investment, and the ability to communicate that trade-off clearly to both engineering and product leadership.
- Familiarity with large-scale ML infrastructure: distributed training (TensorFlow, PyTorch), feature stores, model serving, and online experimentation frameworks.
Qualifications: maybe we could treat the above as Qualifications, but we already used Requirements. Could also have Qualifications separate if there are any explicit qualifications like education? Not present. Could skip Qualifications section if no specific content. But instruction: "Prefer a few substantive sections over many thin ones." We could combine Requirements and Qualifications into one section? But headings are prescribed; we can still include Qualifications if we have content. Since we don't have separate qualifications, maybe we omit Qualifications heading.
Skills: maybe we could extract from Qualities: familiarity with large-scale ML infrastructure etc. Could be Skills. But we already used that in Requirements. Might be redundant. Better to have sections: About the role, Responsibilities, Requirements, Benefits, Pay, Schedule.
Benefits: from Compensation & Benefits: "In addition to salary, you'll be eligible for an equity package, an annual performance bonus, and our competitive benefits that support you and your family." Also maybe mention geographic pay zones? That's Pay.
Pay: Salary Range Geo-1: 296,000.00 - 400,000.00 USD Annual; Geo-2: 266,000.00 - 360,000.00 USD Annual.
Schedule: Location mentions: "We prioritize candidates based within commutable distance of Etsy’s Brooklyn Office. Depending on proximity, we ask team members to come in once or twice per week. Remote candidates outside commutable distance may be considered on a case-by-case basis, with the expectation of occasional travel to the office." That's schedule info (hybrid). So we can put that under Schedule.
We must not include boilerplate like EEO, etc. Also remove "What's Next" and "Our Promise". Also remove "Learn more details about our work modes and workplace safety policies here." (that's a link). Also remove "While Etsy supports visa sponsorship..." etc.
Now produce HTML fragment with headings and appropriate - for lists,
for prose.
We need to ensure we don't include any markdown fences.
Let's craft:
About the role
We are looking for a Director of Engineering, Machine Learning to lead Etsy's Recommendations, the team responsible for the algorithms and systems that power personalized onsite-discovery across our marketplace. This is a pivotal technical leadership role at the intersection of ML research, systems engineering, and product impact. In this role, you will own the end-to-end technical strategy for how Etsy connects buyers with the right listings across the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. You will lead a team of ML engineers, applied scientists, and system engineers organized across three areas: candidate retrieval, ranking, and recommendation systems. You will drive the evolution of our models from state-of-the-art multi-interest retrieval and multi-task deep ranking to next-generation LLM-powered foresight recommendations and generative discovery experiences. The biggest unsolved challenge in front of this team is not just relevance — it is inspiration: helping buyers discover new ideas and start new shopping missions. This opportunity is a full-time position reporting to the VP of Engineering at Search, Recommendations, and Ads.
The Recommendations organization sits at the heart of Etsy's marketplace. We own over 60 recommendation modules across Etsy's surfaces, with our most impactful work powering the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. Our ML stack spans the full recommendation pipeline. On retrieval, we build and evolve our multi-interest user-to-listing embedding model as well as a LLM-powered system that generates forward-looking shopping ideas based on buyer personas. On ranking, we operate a multi-task deep ranking model jointly optimizing for clicks, favorites, ads-to-cart, and purchases. Underpinning all of this is ReactorRecs, our in-house serving and orchestration platform that composes retrieval, ranking, filtering, and experimentation across 200+ recommendation modules. We believe in research grounded in real user impact. We run rigorous offline and online experiments, use LLM-based and human-in-the-loop evaluation, and hold ourselves to both engagement and discovery metrics.
Responsibilities
- Define and drive the multi-year product and technical roadmap for Recommendations ML, with particular focus on personalization, foresight, diversity, and content freshness — the four problems where we have the most room to grow.
- Lead and grow a team of ML engineers, applied scientists, and platform engineers across retrieval, ranking, and systems sub-teams, each with its own engineering manager.
- Partner closely with product, UX, and cross-functional ML teams (Search, Ads, Buyer Understanding, Content Understanding) to align on a coherent, end-to-end buyer experience and shared infrastructure investments.
- Drive the architectural evolution from our current two-stage retrieval-and-ranking pipeline toward more generative and LLM-integrated recommendation architectures.
- Set the technical bar for model development, experimentation, evaluation, and productionization — including our offline VQA and LLM-as-judge evaluation frameworks.
- Champion a high-velocity experimentation culture: our Multi-Variant Dataset (MVD) framework allows parallel A/B tests across retrieval and ranking components, and you should know how to squeeze value out of every experiment.
- Mentor and develop engineering talent at all levels, including Staff and Senior Staff engineers and the engineering managers who report to you.
- Of course, this is just a sample of the kinds of work this role will require! You should assume that your role will encompass other tasks, too, and that your job duties and responsibilities may change from time to time at Etsy's discretion, or otherwise applicable with local law.
Requirements
- Proven ability to lead and develop multi-disciplinary teams of ML engineers, applied scientists, and platform engineers, including managing managers and senior managers.
- 10+ years of progressive experience in ML engineering, with demonstrated success shipping large-scale recommendation, retrieval, or ranking systems into production.
- A track record of driving complex, cross-functional technical programs to completion — including navigating shared infrastructure decisions with partner teams in Search, Ads, or adjacent ML platforms.
- Sharp instincts for balancing
- for lists,
for prose. We need to ensure we don't include any markdown fences. Let's craft:
About the role
We are looking for a Director of Engineering, Machine Learning to lead Etsy's Recommendations, the team responsible for the algorithms and systems that power personalized onsite-discovery across our marketplace. This is a pivotal technical leadership role at the intersection of ML research, systems engineering, and product impact. In this role, you will own the end-to-end technical strategy for how Etsy connects buyers with the right listings across the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. You will lead a team of ML engineers, applied scientists, and system engineers organized across three areas: candidate retrieval, ranking, and recommendation systems. You will drive the evolution of our models from state-of-the-art multi-interest retrieval and multi-task deep ranking to next-generation LLM-powered foresight recommendations and generative discovery experiences. The biggest unsolved challenge in front of this team is not just relevance — it is inspiration: helping buyers discover new ideas and start new shopping missions. This opportunity is a full-time position reporting to the VP of Engineering at Search, Recommendations, and Ads.
The Recommendations organization sits at the heart of Etsy's marketplace. We own over 60 recommendation modules across Etsy's surfaces, with our most impactful work powering the App Homefeed, Web Home, Shop Home, Listing pages, Cart, and Checkout pages. Our ML stack spans the full recommendation pipeline. On retrieval, we build and evolve our multi-interest user-to-listing embedding model as well as a LLM-powered system that generates forward-looking shopping ideas based on buyer personas. On ranking, we operate a multi-task deep ranking model jointly optimizing for clicks, favorites, ads-to-cart, and purchases. Underpinning all of this is ReactorRecs, our in-house serving and orchestration platform that composes retrieval, ranking, filtering, and experimentation across 200+ recommendation modules. We believe in research grounded in real user impact. We run rigorous offline and online experiments, use LLM-based and human-in-the-loop evaluation, and hold ourselves to both engagement and discovery metrics.
Responsibilities
- Define and drive the multi-year product and technical roadmap for Recommendations ML, with particular focus on personalization, foresight, diversity, and content freshness — the four problems where we have the most room to grow.
- Lead and grow a team of ML engineers, applied scientists, and platform engineers across retrieval, ranking, and systems sub-teams, each with its own engineering manager.
- Partner closely with product, UX, and cross-functional ML teams (Search, Ads, Buyer Understanding, Content Understanding) to align on a coherent, end-to-end buyer experience and shared infrastructure investments.
- Drive the architectural evolution from our current two-stage retrieval-and-ranking pipeline toward more generative and LLM-integrated recommendation architectures.
- Set the technical bar for model development, experimentation, evaluation, and productionization — including our offline VQA and LLM-as-judge evaluation frameworks.
- Champion a high-velocity experimentation culture: our Multi-Variant Dataset (MVD) framework allows parallel A/B tests across retrieval and ranking components, and you should know how to squeeze value out of every experiment.
- Mentor and develop engineering talent at all levels, including Staff and Senior Staff engineers and the engineering managers who report to you.
- Of course, this is just a sample of the kinds of work this role will require! You should assume that your role will encompass other tasks, too, and that your job duties and responsibilities may change from time to time at Etsy's discretion, or otherwise applicable with local law.
Requirements
- Proven ability to lead and develop multi-disciplinary teams of ML engineers, applied scientists, and platform engineers, including managing managers and senior managers.
- 10+ years of progressive experience in ML engineering, with demonstrated success shipping large-scale recommendation, retrieval, or ranking systems into production.
- A track record of driving complex, cross-functional technical programs to completion — including navigating shared infrastructure decisions with partner teams in Search, Ads, or adjacent ML platforms.
- Sharp instincts for balancing