Time Series Researcher - Forward deployed

We need to output HTML fragment with sections: About the role? Actually the posting includes "About Applied Computing" then "The Role" then "What You’ll Own", "Must-Have Qualifications", "How We Work", "What This Role Is Not", then bullet points under sections: "Design & Implement Foundational Time Series Models", "Explore and select appropriate objectives", "Develop hybrid approaches", "Embed Physics-Informed Structure", "Uncertainty, Robustness & Reliability", "Production Structured AI Code", "Benchmarking & Validation", "What Success Looks Like". Also there is a concluding paragraph about Applied computing. We need to use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). But we can also use other headings? The instruction: use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). It suggests we should use those headings if they exist. However the posting doesn't have those exact headings; we need to map content to appropriate sections. We should only include sections when posting has real specific content. So we can have sections like "About the role", "Responsibilities", "Requirements", "Qualifications", "Skills", maybe "How We Work" could be part of description but not required heading. We need to decide. We must not repeat job title as heading. No Job ID. We must preserve all substantive information, not invent or summarize away detail. Remove boilerplate (cookie/navigation/legal/EEO etc). There's none. We need to output only HTML fragment. We need to use

• for any list of duties/requirements/benefits, and

  for prose. Thus we need to parse the posting into prose paragraphs and lists. Let's extract sections: - About Applied Computing: description of company. Could be under "About the role"? Actually "About the role" is separate. But we have company overview. Could be a section "About the company" but not in allowed headings. The allowed headings list includes About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. It says use

  for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). It doesn't forbid other headings, but suggests we should use those for sections. Probably we should map content to those headings where applicable. We have: - About Applied Computing: could be considered "About the role"? Not exactly. But maybe we can treat as a brief intro paragraph before sections. - The Role: description of role. Could be under "About the role" heading. - What You’ll Own: responsibilities. Could be under "Responsibilities". - Must-Have Qualifications: qualifications. - How We Work: maybe part of description but could be under "Skills"? Not exactly. Could be omitted or included as prose under responsibilities. - What This Role Is Not: maybe also responsibilities. - Design & Implement Foundational Time Series Models etc: these are more detailed responsibilities/duties. Could be under Responsibilities as list. - What Success Looks Like: maybe under Responsibilities or as separate section? Could be under "Responsibilities" as expectations. - The concluding paragraph about Applied computing: maybe repeat of intro; could be omitted as duplicate? It's somewhat repetitive but we must preserve all substantive information; not summarizing away detail. It says "Applied computing is one of its kind revolution..." This is similar to earlier intro but adds detail. We should keep it. We need to avoid duplication: there may be duplicated lines; we should remove duplicated lines. The concluding paragraph is similar to earlier but not identical; we can keep both? Might be considered duplicated content? It says "Applied computing is one of its kind revolution with a mission to deliver sustainable abundance for a growing planet, through AI that works for the Energy Industry". Earlier we had: "Applied Computing was founded in 2024 to build Orbital, a physics-informed foundation model for energy operations. We’re live across oil and gas, refineries, and petrochemicals, working towards our mission: sustainable abundance for a growing planet. The hydrocarbon industry keeps the world running. But its complexity has left operators tied to legacy systems, making critical decisions on less than 10% of available data. We built Orbital to change that. It’s a foundation model built specifically for energy that lets companies use AI at scale, harnessing all of their operational data and optimising in real time for any metric. Decisions get faster, operations get safer, and carbon intensity falls. We’ve raised over $32 million, including one of the largest seed rounds for an AI company in the UK. We’re just getting started." The concluding paragraph is a subset; it's not exactly duplicate but repeats mission. Might be considered duplicate? The instruction: "Remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions." Duplicated lines: if same sentence appears twice, remove one. The concluding paragraph is not identical lines but similar meaning. Probably we keep both as they add some nuance? Might be considered redundant but not exact duplicate. Safer to keep. Now we need to decide sections. We could have:

  About the role

  The Time Series Researcher owns the core of Orbital’s temporal intelligence. This role exists to design, validate, and deploy foundational time-series models that operate under real world constraints: noisy sensors, partial observability, physical laws, and high economic stakes. This is a customer facing role part of forward deployed team working with customer engineering teams to fine tune our model to their plants. This is not offline research. You will own the full lifecycle; from theoretical formulation and experimentation to real-time inference, uncertainty estimation, and continuous retraining in production.

  Then responsibilities: we can combine "What You’ll Own" and the bullet points under Design & Implement etc. We could have

  Responsibilities

  • Orbital’s foundational time-series modelling stack
  • Physics-informed and probabilistic model design
  • Uncertainty quantification and robustness under sensor faults
  • Research → production translation for time-series models
  • Benchmarking standards and validation protocols used across the company
  Then maybe another list for detailed duties: "Design & Implement Foundational Time Series Models" etc. But we could incorporate them as sub-items? The instruction: use
  • for any list of duties/requirements/benefits. So we can have multiple lists. We need to preserve all substantive information; we can break into multiple
    under same heading? Probably okay. We could have:

    Responsibilities

    • Own Orbital’s foundational time-series modelling stack
    • Design physics-informed and probabilistic models
    • Quantify uncertainty and ensure robustness under sensor faults
    • Translate research to production for time-series models
    • Establish benchmarking standards and validation protocols
    Then maybe another

    Key Activities

    but not allowed heading. Could we use

    Design & Implement Foundational Time Series Models

    ? The instruction says use

    for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). It doesn't forbid other headings but suggests we should use those. Safer to stick to those headings only. So we need to fit all content under those headings. Thus we need to map detailed duties under Responsibilities as list items. We can break responsibilities into multiple list items each containing sub-details maybe using nested lists? Not specified but we can use

    • and inside li we can have

      or further

      . Probably okay. Similarly for Requirements/Qualifications/Skills. We have Must-Have Qualifications list: PhD, first-author publications, 3+ years hands-on research, demonstrated experience in Deep Learning & Probabilistic modelling, Expert Python skills with production-grade PyTorch code, Experience deploying ML models into real systems. We could put those under

      Qualifications

      as list. Skills: maybe Expert Python skills etc could be under Skills, but we already have them in qualifications. Could separate: Qualifications (education, experience) and Skills (technical). But we have limited headings; we can use both. How We Work: maybe part of description but could be under Responsibilities or as a separate section like "How We Work" not in allowed headings. Could incorporate into responsibilities as a paragraph. What This Role Is Not: could be a paragraph under Responsibilities. What Success Looks Like: could be under Responsibilities as expectations. Thus final structure:

      About the role

      ... (role description)

      Responsibilities

      • Own Orbital’s foundational time-series modelling stack
      • Design physics-informed and probabilistic models
      • Quantify uncertainty and ensure robustness under sensor faults
      • Translate research to production for time-series models
      • Establish benchmarking standards and validation protocols
      • Design core time-series architectures supporting forecasting, classification/anomaly detection, optimisation & control-adjacent tasks
      • Explore and select appropriate objectives (probabilistic losses, generative formulations, reinforcement-learning-inspired objectives)
      • Develop hybrid approaches blending classical statistical models, deep learning architectures, physics-based constraints
      • Embed physics-informed structure: integrate conservation laws, process constraints, differential-equation-based priors; improve generalisation, interpretability, extrapolation; ensure physical feasibility in production
      • Design uncertainty-aware models (Bayesian, ensemble, hybrid); quantify confidence under sensor drift/failure, regime change, sparse/delayed ground truth; ensure outputs usable by operations and engineering teams
      • Production structured AI code: containerise and deploy models using Docker on AWS/Azure (EKS, ECS, SageMaker); build/integrate CI/CD pipelines for training, evaluation, rollout/rollback, automated retraining triggers
      • Benchmarking & validation: define rigorous back-testing and evaluation protocols; build automated benchmarking pipelines across datasets, regimes, failure modes; compare against classical baselines and modern deep-learning approaches; ensure claims defensible to customers, partners, internal stakeholders
      • What success looks like: first 90 days – deep understanding of Orbital’s data, domains, production constraints; contribution to at least one core time-series model or experimental track; clear ownership of a modelling problem with defined success metrics. 6–12 months – one or more foundational models running reliably in production; demonstrable improvements in forecast accuracy, robustness under faults, uncertainty calibration; models actively used by downstream agents and optimisation layers; benchmarking standards adopted across research team.

      Qualifications

      • PhD in Computer Science, Statistics, Applied Mathematics, Physics, or related field
      • First-author publications in time-series modelling, forecasting, signal processing, or physics-informed ML
      • 3+ years of hands-on research experience in time-series or sequence modelling
      • Demonstrated experience in Deep Learning & Probabilistic modelling
      • Expert Python skills with production-grade PyTorch code
      • Experience deploying ML models into real systems

      Skills

      Expert Python skills with production-grade PyTorch code (already covered). Could also mention: strong background in physics, statistics, learning as complementary; ability to iterate aggressively, benchmark honestly, ship responsibly; principled models that survive contact with reality.

      But we might duplicate Python skills. Could combine. We also have "How We Work" description: "Research is judged by production impact, not paper count; We value princip