Research Science Intern (PhD)

We need to output HTML fragment with sections for headings: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections that have real content. Must not repeat job title heading. Must not include Job ID etc. Must remove boilerplate like EEO, legal, scam warning, navigation, etc. Must preserve all substantive info. We need to parse the posting. Key content: - Intro: Datadog AI Research — Scholars Program with Carnegie Mellon University. Description: partnering with CMU to support PhD students working on open research problems grounded by ongoing efforts at Datadog/DAIR. You will frame a problem, run experiments, write up, with compute and data at a scale most academic labs cannot provide. You will collaborate with colleagues working on same questions. - The Lab And The Research: description of DAIR: industrial research lab motivated by practical challenges in observability and software operation: detecting and diagnosing failures, understanding complex production environments, helping engineers operate software more effectively. Focus on creating specialized foundation models, post-training and evaluating AI agents, building frontier-scale ML systems. Combining fundamental research with Datadog's large-scale real-world data and infrastructure, lab develops new AI capabilities and translates them into practical systems with meaningful impact. Internship projects are shaped with your DAIR mentor and your CMU faculty advisor. You do not need prior experience with observability, monitoring, or infrastructure. - What You'll Do: list: Own a research project end to end: framing the question, running the experiments, writing it up; Work directly with a DAIR mentor engaged in the same problem, and stay connected to your advisor and lab; Publish, and use the work toward your dissertation; See research reach production, when it works. - Who You Are: Currently enrolled in a PhD program at Carnegie Mellon in machine learning, computer science, statistics, or a related field; Depth in at least one area relevant to the research above; Comfort running real experiments — training models, working with GPUs, reading and reimplementing recent papers; Evidence you can do research: conference or workshop papers, preprints, open-source contributions, or a lab project you can discuss in depth; Research taste. You can explain why a problem matters and what would change if it were solved.; Honest empiricism. You report the ablation that didn't work.; Support from your faculty advisor.; Not Required: A long publication record. How you think matters more than how much you've published.; Observability, monitoring, or SRE background.; A perfect match to the areas above.; Datadog values people from all walks of life. We know not everyone will meet all the above qualifications on day one. That’s okay. If you’re passionate about technology and want to grow your experience, we encourage you to apply. - Additional info: This job is available in various departments within our company; to conform to US export control regulations, some of these roles may require candidates to be eligible for any required authorizations from the US government. If possible, please apply using your personal email address instead of your university email address. Students can sign up for a free Datadog Pro account to learn more about our platform and products. Datadog offers a competitive salary for this role and may include additional compensation elements. Actual compensation is based on factors such as the candidate's skills, qualifications, and experience. In addition, Datadog offers a wide range of best in class, comprehensive and inclusive employee benefits for this role including healthcare, paid time off, and benefits related to traveling and relocation, if eligible. The reasonably estimated yearly salary for this role at Datadog is: $110,000—$140,000 USD. - About Datadog: description of company. - Equal Opportunity etc (boilerplate) to remove. - Privacy and AI Guidelines etc (boilerplate) to remove. We need to produce sections: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. We need to decide what goes where. "About the role": likely the introductory description: Datadog AI Research — Scholars Program with Carnegie Mellon University, description of program, lab, etc. "Responsibilities": What You'll Do bullet list. "Requirements": maybe the "Who You Are" list of required qualifications? Could be split into Requirements and Qualifications. Usually Requirements are mandatory; Qualifications are preferred. The posting lists required: currently enrolled PhD at CMU in ML, CS, stats or related; depth in at least one area; comfort running real experiments; evidence you can do research; research taste; honest empiricism; support from faculty advisor. Not required: long publication record, observability background, perfect match. So we can treat those as Requirements. Qualifications could be "Not Required" items? But we should not create a Qualifications section if it's just "Not Required" filler? The instruction: include a section ONLY when the posting has real, specific content for it — never emit a heading followed by filler/placeholder text. So if we have content for Qualifications (like "Not Required: A long publication record...") that is still substantive? It says not required, but it's still info. Could be considered Qualifications (preferred). However we might combine Requirements and Qualifications into one section? But we have separate headings allowed. We could put Requirements for mandatory items, and Qualifications for preferred/not required items. That seems okay. Skills: maybe list specific skills: training models, working with GPUs, reading and reimplementing recent papers, etc. Could be part of Requirements. Benefits: we have info: Datadog offers a wide range of best in class, comprehensive and inclusive employee benefits for this role including healthcare, paid time off, and benefits related to traveling and relocation, if eligible. Pay: we have salary range: $110,000—$140,000 USD. Schedule: not mentioned. No schedule info (full-time, part-time, etc). So we should NOT include Schedule section. We also need to include maybe a short paragraph for metadata like job ID etc but we should drop. We must not include boilerplate like Equal Opportunity, Privacy, etc. We must not include navigation, cookie, legal, EEO, scam-warning, duplicated lines, application instructions. The posting includes "If possible, please apply using your personal email address instead of your university email address." That's application instruction; we should remove. Also "Students can sign up for a free Datadog Pro account to learn more about our platform and products." That's maybe not application instruction but could be considered extra; but it's not substantive to role? Might be considered extra but could be kept? The instruction: remove application instructions. That line is not exactly instruction to apply but a suggestion. Might be considered extra but could be removed to be safe. Also "This job is available in various departments within our company; to conform to US export control regulations, some of these roles may require candidates to be eligible for any required authorizations from the US government." That's substantive info about eligibility; we should keep maybe in About the role or Requirements? It's about eligibility. Could be placed in Requirements as a note. We need to preserve ALL substantive information; do not invent or summarize away detail. So we need to keep the eligibility note. We need to keep the description of Datadog (About Datadog) maybe as part of About the role? Could be separate but we have only About the role heading; we could include that description there. We must not repeat the job title as a heading. The job title is "Datadog AI Research — Scholars Program with Carnegie Mellon University". We should not put that as heading. We need to output only HTML fragment, no /, no markdown fences. Thus produce something like:

About the role

Datadog AI Research (DAIR) is partnering with Carnegie Mellon University to support a small number of PhD students working on open research problems grounded by ongoing efforts at Datadog/DAIR. You will frame a problem, run your own experiments, and write up what you find, with compute and data at a scale most academic labs cannot provide. You will collaborate with colleagues working on the same questions. DAIR is an industrial research lab motivated by practical challenges in observability and software operation: detecting and diagnosing failures, understanding complex production environments, and helping engineers operate software more effectively. The lab focuses on creating specialized foundation models, post-training and evaluating AI agents, and building frontier-scale machine learning systems. By combining fundamental research with Datadog's large-scale, real-world data and infrastructure, the lab develops new AI capabilities and translates them into practical systems with meaningful impact. Internship projects are shaped with your DAIR mentor and your CMU faculty advisor. You do not need prior experience with observability, monitoring, or infrastructure. This job is available in various departments within our company; to conform to US export control regulations, some of these roles may require candidates to be eligible for any required authorizations from the US government.

Responsibilities

• Own a research project end to end: framing the question, running the experiments, writing it up
• Work directly with a DAIR mentor engaged in the same problem, and stay connected to your advisor and lab
• Publish, and use the work toward your dissertation
• See research reach production, when it works

Requirements

• Currently enrolled in a PhD program at Carnegie Mellon in machine learning, computer science, statistics, or a related field
• Depth in at least one area relevant to the research above
• Comfort running real experiments — training models, working with GPUs, reading and reimplementing recent papers
• Evidence you can do research: conference or workshop papers, preprints, open-source contributions, or a lab project you can discuss in depth
• Research taste: you can explain why a problem matters and what would change if it were solved
• Honest empiricism: you report the ablation that didn't work
• Support from your faculty advisor

Qualifications

• Not required: A long publication record. How you think matters more than how much you've published.
• Observability, monitoring, or SRE background is not required.
• A perfect match to the research areas above is not required.

Skills

Training models, working with GPUs, reading and reimplementing recent papers, conducting research, publishing results.

Benefits

• Healthcare
• Paid time off
• Benefits related to traveling and relocation (if eligible)

Pay

The reasonably estimated yearly salary for this role at Datadog is: $110,000—$140,000 USD.

We should not include Schedule. Check that we didn't include any boilerplate. We removed Equal Opportunity, Privacy, AI Guidelines, etc. We removed application instructions like "If possible, please apply using your personal email address..." and "Students can sign up for a free Datadog Pro account...". We kept the eligibility note. We kept the description of Datadog? Actually we omitted the "About Datadog" paragraph. Should we include it? It's substantive about the company. Could be included in About the role section. The instruction: "Prefer a few substantive sections over many thin ones." We could incorporate that description into About the role. Let's add a paragraph about Datadog after the initial description. Thus About the role paragraph could be longer: include description of DAIR, lab, etc, and also About Datadog. Let's craft:

Datadog AI Research (DAIR) is partnering with Carnegie Mellon University to support a small number of PhD students working on open research problems grounded by ongoing efforts at Datadog/DAIR. You will frame a problem, run your own experiments,