<div dir="ltr"><p dir="LTR" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">The Machine Learning (ML) scientist will apply ML to the Earth system sciences as part of the Analytics and Integrative Machine Learning (AIML) Group in the Technology Development Division (TDD) in the Computational and Information Systems Laboratory (CISL) at the National Center for Atmospheric Research (NCAR). The incumbent will have a proven ability to apply a variety of ML algorithmic approaches to problems in earth systems science or related physical science disciplines.</p><p dir="DEFAULT" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;min-height:16.08px;color:rgb(74,74,74);text-align:inherit;white-space:pre-wrap"></p><p dir="LTR" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">Relevant ML application experience may include the use of ML training and inference systems for the recognition, prediction or tracking of important features or events in datasets, or alternatively, through the auto-encoding of suitable physics parameterizations in earth system models with neural networks, the replacement of model components with efficient, learned emulators. <span style="margin:0px;padding:0px;border:0px;font:inherit;vertical-align:baseline">Machine learning techniques may also be applied by the  scientist to </span>automate or accelerate the human data analysis of hundreds of routine data products, thus amplifying the scientific capability of researchers, the integration of non-traditional data sources into earth system prediction systems, to help optimize supercomputing workflows through ML-guided resource management, or for the early detection and steering of numerical simulations.</p><p dir="DEFAULT" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);text-align:inherit;white-space:pre-wrap"><span class="gmail-WF11" style="margin:0px;padding:0px;border:0px;font:inherit;vertical-align:baseline;word-break:break-all;display:inline;text-decoration:inherit">                                                                                                                                                                                                                  </span></p><h2 style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-weight:500;font-stretch:inherit;font-size:14px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap"><span style="margin:0px;padding:0px;border:0px;font-style:inherit;font-variant:inherit;font-weight:700;font-stretch:inherit;font-size:inherit;line-height:inherit;font-family:inherit;vertical-align:baseline">Responsibilities:</span></h2><p dir="LTR" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">The position works on machine learning projects with scientists, engineers and students in NCAR’s Computational and Information Systems Laboratory (CISL), the Research Applications Laboratory (RAL), the Earth Observing Laboratory (EOL), the High Altitude Observatory (HAO), and the Atmospheric Chemistry, Observations and Modeling (ACOM) Laboratory, and potentially with external data scientists as well. The successful candidate will identify and apply appropriate machine learning techniques focused on two initial projects:</p><p dir="DEFAULT" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;min-height:16.08px;color:rgb(74,74,74);text-align:inherit;white-space:pre-wrap"></p><ul style="margin:12px 0px 0px;padding:0px 40px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;list-style-position:outside;color:rgb(74,74,74);white-space:pre-wrap"><li style="margin:0px;padding:0px;border:0px;font:inherit;vertical-align:baseline"><span style="margin:0px;padding:0px;border:0px;font-style:inherit;font-variant:inherit;font-weight:700;font-stretch:inherit;font-size:inherit;line-height:inherit;font-family:inherit;vertical-align:baseline">Emulating atmospheric organic chemistry reaction networks. </span>Current chemistry-climate models cannot represent the complex chemistry involved in the degradation of hydrocarbons emitted by anthropogenic activities, due to the enormous number of species and reactions. ML emulation may produce less costly reduced models, providing new opportunities to investigate their impact on human health and air quality.</li></ul><p dir="DEFAULT" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;min-height:16.08px;color:rgb(74,74,74);text-align:inherit;white-space:pre-wrap"></p><ul style="margin:12px 0px 0px;padding:0px 40px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;list-style-position:outside;color:rgb(74,74,74);white-space:pre-wrap"><li style="margin:0px;padding:0px;border:0px;font:inherit;vertical-align:baseline"><span style="margin:0px;padding:0px;border:0px;font-style:inherit;font-variant:inherit;font-weight:700;font-stretch:inherit;font-size:inherit;line-height:inherit;font-family:inherit;vertical-align:baseline">Data bottlenecks in cloud observational systems.</span> Estimating the radiation balance in the Earth system is central to predictive climate models and hinges on understanding cloud processes.  The Holographic Detector for Clouds (HOLODEC) is an airborne instrument that gives an unrivaled view of 3-D distributions of droplets, providing an unprecedented accuracy and detail of cloud physics data. However, analyzing the huge data volumes produced by the instrument with current techniques presents a bottleneck, limiting the instrument’s scientific utility. Image-based ML methods could accelerate the analysis process and help advance our understanding of cloud processes. </li></ul><p dir="DEFAULT" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;min-height:16.08px;color:rgb(74,74,74);text-align:inherit;white-space:pre-wrap"></p><p dir="LTR" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">For each project, the scientist will work with the team, and in collaboration with domain scientists, to create and share the necessary training datasets, and apply, tune and evaluate and verify a variety of machine learning approaches to solving these problems. The machine learning scientist’s efforts will be built on top of NCAR’s core capability in domain-focused statistical development, and will leverage its vast observational and model output datasets, and CISL’s petascale supercomputing infrastructure, and cloud-based resources and environments. The position will require the ability to work in teams and across disciplines in order to cross-fertilize ideas and build strong collaborations to tackle Earth system science challenges. This integration with and support from colleagues in the earth system sciences will help to ensure the relevance and sustainability of the ML project scientist’s research activities.</p><p dir="DEFAULT" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;min-height:16.08px;color:rgb(74,74,74);text-align:inherit;white-space:pre-wrap"></p><p dir="LTR" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">The position provides high-level machine learning expertise to these projects, assists in planning the projects human and financial resource requirements, and will participate in the evaluation of the project’s progress, its results, and make adjustments to the project’s approach to better achieve objectives. The Machine Learning scientist may also serve, from time to time, as a consultant to internal staff and external organizations on machine learning topics.</p><p dir="LTR" style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap"><br></p><p style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">See the full job ad and apply at <a href="https://ucar.wd5.myworkdayjobs.com/en-US/UCAR_Careers/job/Mesa-Lab-Building/Machine-Learning-Scientist_REQ-2019-39-1" style="font-family:Arial,Helvetica,sans-serif;font-size:small">https://ucar.wd5.myworkdayjobs.com/en-US/UCAR_Careers/job/Mesa-Lab-Building/Machine-Learning-Scientist_REQ-2019-39-1</a>. </p><p style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap"><br></p><p style="margin:0px;padding:0px;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-stretch:inherit;font-size:12px;line-height:inherit;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif;vertical-align:baseline;color:rgb(74,74,74);white-space:pre-wrap">David John Gagne</p></div>