Research Intern, Scene Understanding and Planning

Job description

<h3 data-anchor="6cc4c0b6-0d6e-4481-8ad8-d8f86b1375b9" data-collapse="false" data-pm-slice="1 3 []"><span data-author="shiyingzhou">About Didi Autonomous Driving</span></h3>
<p data-anchor="e9a36b25-cf0d-4bb1-918b-c632d766576d"><span data-author="shiyingzhou">Founded in 2016, DiDi's autonomous driving unit is dedicated to developing Level 4 autonomous driving technology to make transportation safer, more efficient, and accessible. In 2019, the unit became an independent company, Voyager, to accelerate technology development, commercialization, and large-scale deployment. By combining DiDi's unique data advantage, operational expertise, and mobility ecosystem, we aim to build a reliable, scalable autonomous fleet that integrates seamlessly with human-driven services. <strong>Driven by real-world customer needs and a strong commitment to safety and user experience</strong>, our mission is to deliver autonomous mobility solutions that create meaningful value for riders and communities.</span></p>
<h3 class="ql-direction-ltr ql-long-10000202573"> </h3>
<h3 class="ql-direction-ltr ql-long-10000202573"><strong class="ql-author-10000202573 ql-size-12 ql-font-arial">About the Role</strong></h3>
<p data-anchor="1a72f168-1c3d-45d5-aca5-33d189d89cf9"><span data-author="yiyang1">We want our vehicles to understand the physical world as it is right now, instead of relying on a map drawn long ago. Roads change, and the world a car actually drives in is always more complex than any static record of it. We are seeking a Research Intern to help give our planner a m</span><span data-author="runyugao">ore accurate and</span><span data-author="yiyang1"> up-to-date picture of the world, so the vehicle handles complex real-world scenes more robustly. In this role, you will take on hard, open problems from real-world driving and solve them with frontier research methods. You will report directly to the team lead and work on the problem from definition to validated solution. Strong results will often be worth publishing, and we will support you in writing them up, but the goal is to solve problems that matter to the product.</span></p>
<h3 data-anchor="a334eacb-e50e-40cf-8cc1-c17d38dd6b94" data-collapse="false"> </h3>
<h3 data-anchor="a334eacb-e50e-40cf-8cc1-c17d38dd6b94" data-collapse="false"><span data-author="yiyang1">Responsibilities</span></h3>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="190f47ec-1d77-43cc-865d-32fbc89224f5">
<li data-anchor="2cbeea4d-ffed-42e3-849c-de863b672e3b">
<p data-anchor="6bc0f23b-cb62-4c96-85da-bea799c90456"><span data-author="yiyang1"> Survey the latest research in scene understanding, prediction, and planning, and turn it into practical approaches that fit our system and constraints.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="244e9f2c-52c0-4755-a78d-cb1667948a68">
<li data-anchor="e99c2e12-ef10-4988-b2a7-6f4790c37526">
<p data-anchor="fa7b21f2-4c9b-42dd-9dec-7ff0d81d4501"><span data-author="yiyang1"> Design and implement deep learning models that build a richer, real-time understanding of the driving environment from onboard sensors and other available sources.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="a7edea2e-6aaa-4967-a874-75df4eb9cb5e">
<li data-anchor="0febfc17-2690-4999-9a76-6bbf2efef250">
<p data-anchor="1097550a-fdf4-40aa-8291-e320f53251ab"><span data-author="yiyang1"> Train and evaluate models on large-scale data from our driving logs and on public benchmarks, and lead failure analysis to find out where and why models fall short.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="98d82dbe-0990-4168-a279-218f54cca193">
<li data-anchor="135e0cea-6bb2-4125-9cc9-b1cf014e6f57">
<p data-anchor="a48b5e43-5e53-4e9b-9f95-139833fddba4"><span data-author="yiyang1"> Define metrics and evaluation methods that connect perception quality to how the vehicle actually drives.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="e6f91098-0ca3-4b7d-b26a-ed9c756f1763">
<li data-anchor="fa9ae119-9980-41df-b72c-afcf078e9ee9">
<p data-anchor="65a6c833-43e0-41d8-b0f9-32187c055ee8"><span data-author="yiyang1"> Turn successful prototypes into solutions that work with our planning stack, with help from production engineers.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="d3bd1420-ab54-419d-b902-eb88f334367a">
<li data-anchor="8756884e-9bcc-4d3d-9883-b4f9720f32de">
<p data-anchor="d5f4374f-876c-40ec-833f-cf9b62e28201"><span data-author="yiyang1"> Collaborate with Perception, Prediction, Planning, and Mapping teams so results move from experiment to product.</span></p>
</li>
</ul>
<h2 data-anchor="22b8e6cc-c83a-4989-8a5e-2161c70fa5af" data-collapse="false"><span data-author="yiyang1">  </span></h2>
<h3 data-anchor="22b8e6cc-c83a-4989-8a5e-2161c70fa5af" data-collapse="false"><span data-author="yiyang1">Qualifications</span></h3>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="e55b95f8-ac03-401a-b26a-dd716e6e560e">
<li data-anchor="b64ceb43-42bf-4eec-b9d3-0f90c0c6084b">
<p data-anchor="3dab7fb3-f1f2-4aa4-8a59-29818b0f98b8"><span data-author="yiyang1"> Currently enrolled in an M.S. or Ph.D. program in Computer Science, Robotics, Electrical Engineering, Machine Learning, or a related field, for the full duration of the internship.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="44492d3e-f03c-49b6-9304-6d1c4d085926">
<li data-anchor="65090bd5-efb5-488d-9a99-0759577aa1e1">
<p data-anchor="df31f0e3-41a9-4f3a-8287-36ab9c4dddf8"><span data-author="yiyang1"> Available for a full-time onsite internship of 3–6 months; start dates are flexible.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="4d88c079-0099-4401-82e7-4ba8873b7bfd">
<li data-anchor="756a9f7a-2169-4d60-baa2-23772f63aa57">
<p data-anchor="65db1356-c8e9-410b-8824-ec4a98ad5e24"><span data-author="yiyang1"> Strong programming skills in Python, and experience with modern deep learning frameworks such as PyTorch or JAX.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="44f3508a-1b2c-4e8b-9441-c4db31a6cf95">
<li data-anchor="8040d6ca-8793-4e5a-9100-34487c74257f">
<p data-anchor="209d2ab4-3f71-4dde-9541-a51fee90012b"><span data-author="yiyang1"> Hands-on experience designing, training, and debugging deep learning models for computer vision, 3D perception, or sequential decision-making.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="5ffbe83c-d7c8-4741-a190-8fed2d3319d9">
<li data-anchor="91c281e2-1418-4a36-aef8-947488f1bd91">
<p data-anchor="36405377-4b84-4c85-971d-8a870393eee3"><span data-author="yiyang1"> Solid understanding of autonomous driving fundamentals, including perception, prediction, and motion planning.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="7d0594ac-46ea-4888-9a31-5578328c01c7">
<li data-anchor="929e96fe-dbd8-42be-8774-c0171975f3cb">
<p data-anchor="2978933d-8a76-4b6f-86af-3ca9a7a1fa0f"><span data-author="yiyang1"> Ability to read research critically, form hypotheses, and turn them into well-controlled experiments.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="04c9c89c-d812-4a05-ac40-fb1f53d354e0">
<li data-anchor="ec86b04d-f748-40e9-91f7-14af81972307">
<p data-anchor="9232a68d-090c-4253-8ddb-3be2a9afa900"><span data-author="yiyang1"> Excellent communication and collaboration skills.</span></p>
</li>
</ul>
<h2 data-anchor="bf6ea757-6c66-47f9-aa9d-7f2170cfe58e" data-collapse="false"><span data-author="yiyang1">  </span></h2>
<h3 data-anchor="bf6ea757-6c66-47f9-aa9d-7f2170cfe58e" data-collapse="false"><span data-author="yiyang1">Preferred Qualifications</span></h3>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="fa5778c5-b2a2-4e18-8359-cd5f58ec0309">
<li data-anchor="04b23951-b818-4a35-88cf-1161d3d06086">
<p data-anchor="3482c247-c810-4e74-af6d-a7cdbbe92a05"><span data-author="yiyang1">Publications at top venues such as CVPR, ICCV, ECCV, NeurIPS, ICLR,</span><span data-author="runyugao"> ICML,</span><span data-author="yiyang1"> ICRA, IROS,</span><span data-author="runyugao"> CoRL</span><span data-author="yiyang1"> or IV.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="1eecb008-f899-484d-95c5-6c814a61a575">
<li data-anchor="f0661f41-0703-4240-94f7-9fec3080ba7c">
<p data-anchor="f5236bbf-989e-4149-b73c-1ae11849d9ad"><span data-author="yiyang1">Experience with multi-modal sensor fusion (camera, lidar, radar) or BEV representations.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="a1a58d3c-4611-4777-b7aa-e208a42a07ab">
<li data-anchor="fb3ab783-2e9e-4a39-9996-f2bbacbe80ff">
<p data-anchor="efce1603-6f29-4176-9e0a-aeef2a8446ff"><span data-author="yiyang1">Experience with transformers, graph neural networks, or other structured scene representations.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="eb18812f-b329-4a1d-91e9-b127e00b2ca7">
<li data-anchor="5014c102-6b34-414a-88b1-d572f136ff71">
<p data-anchor="3e3d20b2-dd4b-4856-8298-6f94f59cecd8"><span data-author="yiyang1">Experience with uncertainty estimation, or with learning-based prediction or planning.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="82e4699c-b093-45f5-ab76-2021d1624543">
<li data-anchor="070fea63-3ae6-4d2e-89ff-17cd49ab2b0c">
<p data-anchor="c359e09b-d48b-4240-acf0-224e830d9cde"><span data-author="yiyang1">Experience with large-scale model training and data pipelines.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="2796720b-5aee-402b-9086-7fa777894dd5">
<li data-anchor="52122d38-bb85-40f9-9db5-66798e7d266d">
<p data-anchor="28ee36e7-abd2-4b39-8fad-69ada49fdfbc"><span data-author="yiyang1">Experience with public autonomous driving datasets such as nuScenes, Argoverse 2, or the Waymo Open Dataset.</span></p>
</li>
</ul>
<ul class="didoc2-editor-bullet-list" type="disc" data-anchor="1fbb4f81-457a-43bd-a7e0-e386b5434c24">
<li data-anchor="339b7e84-3ea9-4b27-a198-63a16c8482c2">
<p data-anchor="2a766215-f7d9-49c0-83ac-5c8cd65aba3b"><span data-author="yiyang1">Proficiency in C++ (CUDA is a plus) and experience with robotics systems.</span></p>
</li>
</ul>
<p class="ql-direction-ltr ql-long-10000202573"> </p>
<p class="ql-direction-ltr ql-long-10000202573"><span class="ql-author-10000202573 ql-size-12 ql-font-arial">The hourly rate for the Intern position in the selected city is $46.</span></p>
<p class="ql-direction-ltr ql-long-10000202573"><em class="ql-author-10000202573 ql-font-arial">I acknowledge that prior to submitting this application, I have read and accepted the Privacy Notice for California Residents which is available on </em><a class="ql-link ql-author-10000202573 ql-font-arial" href="https://v.didi.cn/AQnxlBa" target="_blank"><em>https://v.didi.cn/AQnxlBa</em></a></p>