Senior Staff Software Engineer, Experimentation Platform

<div class="content-intro"><p><img style="display: none; max-width: 100%;" src="https://click.appcast.io/greenhouse-te8/a31.png?ent=34e=22630t=1701374353806" width="1px"> <img style="display: none; max-width: 100%;" src="https://track.jobadx.com/v1/i.gif?utm_pixel=224e990b-8ff4-4287-8d5d-2ff09647f181utm_ptz=ESTutm_rqt=track" alt="" width="1"></p></div><h1><strong>About the Team</strong></h1> <p>The Experimentation Platform team develops an industry-leading platform that enables data scientists, ML engineers and non-technical users to design, run and analyze experiments and conduct exploratory and causal analysis. At DoorDash, where we run thousands of experiments per month, our mission is to equip all decision makers with rigorous, data-driven insights by democratizing experimentation with quality and velocity. The team consists of a mix of experienced veterans of backend, web, statistical and data infra engineers that works closely with the data science community. </p> <p>Some of the interesting work done in the team was published in articles such as:(1) <a href="https://doordash.engineering/2022/05/24/meet-dash-ab-the-statistics-engine-of-experimentation-at-doordash/">Meet Dash-AB-The Statistics Engine of Experimentation at DoorDash</a>, (2) <a href="https://doordash.engineering/2020/09/09/experimentation-analysis-platform-mvp/">Supporting Rapid Product Iteration with an Experimentation Analysis Platform</a>, (3) <a href="https://careersatdoordash.com/blog/experiment-faster-and-with-less-effort/">Experiment Faster and with Less Effort</a>, (4)<a href="https://careersatdoordash.com/blog/doordash-experimentation-with-interleaving-designs/"> Interleaving designs</a>, (5) <a href="https://arxiv.org/abs/2311.14698">Fractional Factorial Design for Business Policy</a>. We help enable and unlock interesting solutions our product teams use such as (3) <a href="https://doordash.engineering/2021/09/21/the-4-principles-doordash-used-to-increase-its-logistics-experiment-capacity-by-1000/">The 4 Principles DoorDash Used to Increase Its Logistics Experiment Capacity by 1000%</a>, (4) <a href="https://doordash.engineering/2020/10/07/improving-experiment-capacity-by-4x/">Improving Online Experiment Capacity by 4X with Parallelization and Increased Sensitivity</a>, etc.</p> <h1><strong>About the Role</strong></h1> <p>If you embrace the challenges in the intersection of statistics, machine learning, and engineering, build cutting-edge experimentation algorithms and work with some of the smartest people in the industry, DoorDash’s Experimentation Platform is the right place for you. Come join us and be part of the mission.</p> <p>You will partner with backend and frontend engineers and build platforms to power data-driven decision making. Solutions will inform how we build our intelligent, last-mile delivery platform for local cities and will be on the forefront of research conducted on three-sided marketplaces.</p> <h1><strong>You're excited about this opportunity because you will...</strong></h1> <ul> <li>Build Experimentation platform that can evolve to handle new statistical methodologies, machine learning and artificial intelligence technologies and advanced causal inference and data mining techniques</li> <li>Drive the statistical and ML development of internal platforms, including both the theoretical and engineering aspects, products including A/B testing platform, Causal Inference platform and Adaptive Learning platform (RL/MAB)</li> <li>Expand the statistical and causal inference algorithms to support large-scale experimentation volume and computation load and high noise-to-signal business environment</li> <li>Apply semi-supervised learning, LLM, active learning, documentation embedding/retrieval and data augmentation strategies to advance the hypothesis generation of the experimentation platform</li> <li>Advise data scientists, operators, and engineers across the company on experimental design and adoption of experimentation tools</l ... (truncated, view full listing at source)