Staff Machine Learning Engineer - Ads Bidding
Job Description
<div class="content-intro"><p><strong>Our Global Structure</strong></p> <p><br>Deliveroo is now part of DoorDash, bringing together teams with even greater reach, scale, and ambition. Depending on your role, you may collaborate with teammates, systems, and leaders across DoorDash and Wolt. Together, we’re unlocking new possibilities as one global team.</p></div><h1><span style="font-size: 14pt; font-family: helvetica, arial, sans-serif;"><strong>Staff Machine Learning Engineer – Ads Bidding</strong></span></h1> <h2><span style="font-size: 14pt; font-family: helvetica, arial, sans-serif;"><strong>The Team</strong></span></h2> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">The Ads ML team builds and maintains the machine learning models powering Deliveroo's advertising product — connecting consumers with relevant ads and helping merchants, grocers, and FMCG brands grow through the platform. The UK team is ~11 MLEs across ads and promotions, and collaborates closely with the US team (meetings capped at 4pm UK to protect reasonable hours).</span></p> <h2><span style="font-size: 14pt; font-family: helvetica, arial, sans-serif;"><strong>The Problem</strong></span></h2> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Deliveroo is migrating ads from cost-per-click (CPC) to cost-per-order (CPO) — merchants pay when an ad drives an order, not when it's clicked. This is a genuinely open, cutting-edge problem: very few companies have solved it, and there's no established playbook to follow. The exact approach is under active discussion and you'd be central to driving it.</span></p> <h2><span style="font-size: 14pt; font-family: helvetica, arial, sans-serif;"><strong>You will:</strong></span></h2> <ul> <li style="font-family: helvetica, arial, sans-serif;"> <p><span style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Own the design and execution of the CPC-to-CPO migration, including the core open question of whether the system stays CPO-facing externally while optimising CPC behind the scenes, or some hybrid of the two</span></p> </li> <li style="font-family: helvetica, arial, sans-serif; font-size: 12pt;"> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Build and improve the bidding surface: goal-based automated bidding, Target CPO, optimised CPC, and Target ROAS</span></p> </li> <li style="font-family: helvetica, arial, sans-serif; font-size: 12pt;"> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Own yield management tied to ROS targeting</span></p> </li> <li style="font-family: helvetica, arial, sans-serif; font-size: 12pt;"> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Partner daily with the DoorDash US team to align on approach across markets</span></p> </li> <li style="font-family: helvetica, arial, sans-serif; font-size: 12pt;"> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Set technical direction for the domain</span></p> </li> </ul> <h2><span style="font-size: 14pt; font-family: helvetica, arial, sans-serif;"><strong>Requirements</strong></span></h2> <ul> <li style="font-family: helvetica, arial, sans-serif;"> <p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Expertise in ad economics, marketplace bidding, or auction theory</span></p> </li> <li style="font-family: helvetica, arial, sans-serif; fo