Ruth Mottram<p>Next week our ECR group are starting a <a href="https://fediscience.org/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> sprint to speed up the development of our new <a href="https://fediscience.org/tags/IceSheet" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>IceSheet</span></a> surface mass budget emulation models. </p><p>I'm looking for hints, ideas tips and inspiration to help smooth and accelerate the process. </p><p>If you're used to working in <a href="https://fediscience.org/tags/Sprint" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Sprint</span></a> mode, what in your opinion makes it work or otherwise hinders it? </p><p>What can I as supervisor do to make it run better? <br>And how should we celebrate it's conclusion? </p><p>Also, should me do an <a href="https://fediscience.org/tags/AMA" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AMA</span></a> on the process? </p><p><a href="https://fediscience.org/tags/ClimateScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ClimateScience</span></a> <a href="https://fediscience.org/tags/IceSheets" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>IceSheets</span></a> <a href="https://fediscience.org/tags/Greenland" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Greenland</span></a> <a href="https://fediscience.org/tags/Antarctica" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Antarctica</span></a></p>