Senior Context Fusion AI Engineer - Autonomous Vehicles
for sections like About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections where posting has real specific content. We need to parse the posting. Job title: not needed as heading. We have sections: "What You'll Be Doing" => Responsibilities. "What We Need To See" => Requirements/Qualifications. "Ways To Stand Out From The Crowd" => Skills (or maybe Qualifications). Could treat as Skills. Pay: there is salary range info: "Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 184,000 USD - 287,500 USD for Level 4, and 224,000 USD - 356,500 USD for Level 5. You will also be eligible for equity and benefits." Thus we have Pay and Benefits sections. Benefits: mention equity and benefits; but not specific details. However we have "You will also be eligible for equity and benefits." That's vague but still content about benefits. Should we include Benefits section? The rule: include a section ONLY when posting has real, specific content for it — never emit a heading followed by filler/placeholder text (e.g. "commensurate with experience", "not specified", "N/A", "varies", "TBD") and never create Pay/Schedule/Benefits sections unless the posting actually states them. Here we have specific salary ranges, so Pay section is valid. Benefits: "eligible for equity and benefits" is vague but still states benefits exist. Might be considered specific enough? It says equity and benefits, but not details. Might be considered filler? The rule says avoid filler like "commensurate with experience", "not specified", "N/A", "varies", "TBD". "eligible for equity and benefits" is not exactly filler; it's a statement that benefits exist. I'd include Benefits section with that sentence. Schedule: not mentioned (full time is given in metadata but we should not put Job ID etc. Full time is mentioned in metadata line "Time Type Full time". Should we include Schedule? The posting says "Time Type Full time". That's schedule info. Should we include Schedule section? The rule: include a section ONLY when posting has real, specific content for it. It does state full time. So we could include Schedule with that. However we must not repeat job title as heading, and not put Job ID/requisition numbers as headings — fold minor metadata into a short
or drop it. The "Time Type Full time" is metadata; we could incorporate into a short
maybe under About the role? But we could also make a Schedule section with that content. The rule says prefer a few substantive sections over many thin ones. Might be better to incorporate schedule into About the role paragraph. We need About the role section? The posting includes introductory paragraph: "We are looking for a strong engineer to join the DRIVE Road Structure / Online Mapping / Context Fusion team. In this role, you will help craft and guide the future of our L3/L4 autonomous-driving solution by building a complete, learned 3D/4D world model that fuses navigation, ego-motion, perception, and sensor signals. You will work closely with perception, prediction, planning, and simulation teams to deliver a world representation that is complete, temporally consistent, uncertainty-aware, and robust enough to drive through the most challenging roads and intersections in L3/L4 autonomy level. This role is central to our vision for AV: developing a shared multimodal scene representation that can jointly support road understanding, 3D object and occupancy perception, motion prediction, and route-conditioned planning. Are you interested in inventing human-level AI for navigation in the unconstrained world under any conditions? If so, join us!" That's suitable for About the role. Thus sections: About the role, Responsibilities, Requirements (or Qualifications), Skills, Pay, Benefits. Possibly also Schedule but we can incorporate. We must not repeat job title as heading. Use
for each section heading.
We need to use - for any list of duties/requirements/benefits, and
for prose.
We need to preserve ALL substantive information; do not invent or summarize away detail.
We need to remove boilerplate: cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. At end there is EEO statement etc. Remove that. Also remove "Applications for this job will be accepted at least until September 22, 2026." That's application instruction; remove. Also remove "This posting is for an existing vacancy." maybe minor metadata; could drop. Also remove "NVIDIA uses AI tools in its recruiting processes." maybe boilerplate; remove. Also remove the hashtags "#AutonomousVehicles #AutonomousVehicles". Remove.
Also remove "We believe that building self-driving vehicles will be a defining contribution of our generation (e.g. traffic accidents are responsible for ~1.25 million deaths per year world-wide). We have the funding and scale, but we need your help on our team. NVIDIA is widely considered to be one of the technology world’s most desirable employers with some of the most forward-thinking people in the world working here. If you're entrepreneurial and autonomous, we want to hear from you!" This is somewhat promotional but still substantive? Might be considered boilerplate? It's not a standard EEO; it's motivational. Could keep? The instruction says remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. This is not exactly those; it's more company pitch. Might keep as part of About the role? But it's after the "What We Need To See" etc. It's more like a closing motivational paragraph. Could keep as part of About the role or maybe a separate section? But we already have About the role covering initial intro. Could append that paragraph to About the role as additional prose.
We need to ensure we don't duplicate lines.
Let's extract content.
**About the role** (intro + motivational paragraph). We'll combine:
"We are looking for a strong engineer to join the DRIVE Road Structure / Online Mapping / Context Fusion team. In this role, you will help craft and guide the future of our L3/L4 autonomous-driving solution by building a complete, learned 3D/4D world model that fuses navigation, ego-motion, perception, and sensor signals. You will work closely with perception, prediction, planning, and simulation teams to deliver a world representation that is complete, temporally consistent, uncertainty-aware, and robust enough to drive through the most challenging roads and intersections in L3/L4 autonomy level. This role is central to our vision for AV: developing a shared multimodal scene representation that can jointly support road understanding, 3D object and occupancy perception, motion prediction, and route-conditioned planning. Are you interested in inventing human-level AI for navigation in the unconstrained world under any conditions? If so, join us! We believe that building self-driving vehicles will be a defining contribution of our generation (e.g. traffic accidents are responsible for ~1.25 million deaths per year world-wide). We have the funding and scale, but we need your help on our team. NVIDIA is widely considered to be one of the technology world’s most desirable employers with some of the most forward-thinking people in the world working here. If you're entrepreneurial and autonomous, we want to hear from you!"
That seems fine.
**Responsibilities** (What You'll Be Doing). Need to convert bullet-like sentences into
- . The paragraph contains multiple sentences separated by periods and semicolons. We'll split into logical items.
Original:
"Design and develop learning-based, multimodal sensor-fusion systems that transform synchronized sensor history, ego-motion, navigation context, and driving context into a unified spatiotemporal world representation.Build architectures that jointly reason over camera, LiDAR, radar, and vehicle-state inputs, with appropriate handling of calibration, synchronization, coordinate transforms, sensor latency, and uncertainty.Develop end-to-end and multi-task models that produce driving-relevant outputs from a shared scene representation, including; road graph elements such as lanes, boundaries, crosswalks, and traffic controls; semantic scene understanding; occupancy and free-space representations, including uncertain and occluded regions;Develop scalable multimodal fusion architectures, including Transformer-based early, late, and hierarchical fusion; BEV, point/voxel, and image-based representations; temporal context aggregation; and cross-modal attention.Create training, fine-tuning, and evaluation pipelines for large-scale multimodal datasets. Define multi-task objectives and metrics that balance perception quality, geometric consistency, prediction accuracy, latency, and safety-critical behavior.Investigate foundation-model approaches for autonomous driving, including vision-language models, multimodal pre-training, representation learning, and efficient deployment of learned world models.Work closely with perception, mapping, prediction, planning, simulation, data, and embedded-software teams to convert research advances into robust, production-quality AV systems.Develop systematic analysis and debugging tools for model failures, cross-sensor disagreement, long-tail scenarios, distribution shift, and regressions in closed-loop simulation and on-road evaluation."
We need to split into list items. Each sentence seems to be a responsibility. Let's create items:
- Design and develop learning-based, multimodal sensor-fusion systems that transform synchronized sensor history, ego-motion, navigation context, and driving context into a unified spatiotemporal world representation.
- Build architectures that jointly reason over camera, LiDAR, radar, and vehicle-state inputs, with appropriate handling of calibration, synchronization, coordinate transforms, sensor latency, and uncertainty.
- Develop end-to-end and multi-task models that produce driving-relevant outputs from a shared scene representation, including road graph elements such as lanes, boundaries, crosswalks, and traffic controls; semantic scene understanding; occupancy and free-space representations, including uncertain and occluded regions.
- Develop scalable multimodal fusion architectures, including Transformer-based early, late, and hierarchical fusion; BEV, point/voxel, and image-based representations; temporal context aggregation; and cross-modal attention.
- Create training, fine-tuning, and evaluation pipelines for large-scale multimodal datasets.
- Define multi-task objectives and metrics that balance perception quality, geometric consistency, prediction accuracy, latency, and safety-critical behavior.
- Investigate foundation-model approaches for autonomous driving, including vision-language models, multimodal pre-training, representation learning, and efficient deployment of learned world models.
- Work closely with perception, mapping, prediction, planning, simulation, data, and embedded-software teams to convert research advances into robust, production-quality AV systems.
- Develop systematic analysis and debugging tools for model failures, cross-sensor disagreement, long-tail scenarios, distribution shift, and regressions in closed-loop simulation and on-road evaluation.
Check for any missing: The original had "including;" after "including;" maybe a typo; we handled.
**Requirements / Qualifications** (What We Need To See). The paragraph:
"What We Need To See BS, MS, or PhD in Computer Science, Robotics, Electrical Engineering, Machine Learning, or a related technical field, or equivalent experience.8+ years of experience, with at least 2+ years in the AV or robotics industry and 2+ years of leadership experience in a technically areaStrong experience developing production-quality sensor-fusion, perception, state-estimation, or autonomous-driving systems.Demonstrated experience with learning-based multimodal perception or fusion involving two or more cameras, LiDAR, radar, map, navigation, and ego-motion signals.Solid understanding of 3D geometry, coordinate frames, calibration, temporal synchronization, ego-motion compensation, tracking, uncertainty estimation, and sensor failure modes.Experience with deep-learning methods for 3D perception, lego-context scene representation, occupancy/occlusion prediction, semantic segmentation, object detection/tracking, motion prediction, or planning.Strong C++ and Python programming skills, with hands-on experience developing, training, and optimizing deep-learning models in PyTorch. Experience with CUDA
for prose. We need to preserve ALL substantive information; do not invent or summarize away detail. We need to remove boilerplate: cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. At end there is EEO statement etc. Remove that. Also remove "Applications for this job will be accepted at least until September 22, 2026." That's application instruction; remove. Also remove "This posting is for an existing vacancy." maybe minor metadata; could drop. Also remove "NVIDIA uses AI tools in its recruiting processes." maybe boilerplate; remove. Also remove the hashtags "#AutonomousVehicles #AutonomousVehicles". Remove. Also remove "We believe that building self-driving vehicles will be a defining contribution of our generation (e.g. traffic accidents are responsible for ~1.25 million deaths per year world-wide). We have the funding and scale, but we need your help on our team. NVIDIA is widely considered to be one of the technology world’s most desirable employers with some of the most forward-thinking people in the world working here. If you're entrepreneurial and autonomous, we want to hear from you!" This is somewhat promotional but still substantive? Might be considered boilerplate? It's not a standard EEO; it's motivational. Could keep? The instruction says remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. This is not exactly those; it's more company pitch. Might keep as part of About the role? But it's after the "What We Need To See" etc. It's more like a closing motivational paragraph. Could keep as part of About the role or maybe a separate section? But we already have About the role covering initial intro. Could append that paragraph to About the role as additional prose. We need to ensure we don't duplicate lines. Let's extract content. **About the role** (intro + motivational paragraph). We'll combine: "We are looking for a strong engineer to join the DRIVE Road Structure / Online Mapping / Context Fusion team. In this role, you will help craft and guide the future of our L3/L4 autonomous-driving solution by building a complete, learned 3D/4D world model that fuses navigation, ego-motion, perception, and sensor signals. You will work closely with perception, prediction, planning, and simulation teams to deliver a world representation that is complete, temporally consistent, uncertainty-aware, and robust enough to drive through the most challenging roads and intersections in L3/L4 autonomy level. This role is central to our vision for AV: developing a shared multimodal scene representation that can jointly support road understanding, 3D object and occupancy perception, motion prediction, and route-conditioned planning. Are you interested in inventing human-level AI for navigation in the unconstrained world under any conditions? If so, join us! We believe that building self-driving vehicles will be a defining contribution of our generation (e.g. traffic accidents are responsible for ~1.25 million deaths per year world-wide). We have the funding and scale, but we need your help on our team. NVIDIA is widely considered to be one of the technology world’s most desirable employers with some of the most forward-thinking people in the world working here. If you're entrepreneurial and autonomous, we want to hear from you!" That seems fine. **Responsibilities** (What You'll Be Doing). Need to convert bullet-like sentences into
- . The paragraph contains multiple sentences separated by periods and semicolons. We'll split into logical items. Original: "Design and develop learning-based, multimodal sensor-fusion systems that transform synchronized sensor history, ego-motion, navigation context, and driving context into a unified spatiotemporal world representation.Build architectures that jointly reason over camera, LiDAR, radar, and vehicle-state inputs, with appropriate handling of calibration, synchronization, coordinate transforms, sensor latency, and uncertainty.Develop end-to-end and multi-task models that produce driving-relevant outputs from a shared scene representation, including; road graph elements such as lanes, boundaries, crosswalks, and traffic controls; semantic scene understanding; occupancy and free-space representations, including uncertain and occluded regions;Develop scalable multimodal fusion architectures, including Transformer-based early, late, and hierarchical fusion; BEV, point/voxel, and image-based representations; temporal context aggregation; and cross-modal attention.Create training, fine-tuning, and evaluation pipelines for large-scale multimodal datasets. Define multi-task objectives and metrics that balance perception quality, geometric consistency, prediction accuracy, latency, and safety-critical behavior.Investigate foundation-model approaches for autonomous driving, including vision-language models, multimodal pre-training, representation learning, and efficient deployment of learned world models.Work closely with perception, mapping, prediction, planning, simulation, data, and embedded-software teams to convert research advances into robust, production-quality AV systems.Develop systematic analysis and debugging tools for model failures, cross-sensor disagreement, long-tail scenarios, distribution shift, and regressions in closed-loop simulation and on-road evaluation." We need to split into list items. Each sentence seems to be a responsibility. Let's create items: - Design and develop learning-based, multimodal sensor-fusion systems that transform synchronized sensor history, ego-motion, navigation context, and driving context into a unified spatiotemporal world representation. - Build architectures that jointly reason over camera, LiDAR, radar, and vehicle-state inputs, with appropriate handling of calibration, synchronization, coordinate transforms, sensor latency, and uncertainty. - Develop end-to-end and multi-task models that produce driving-relevant outputs from a shared scene representation, including road graph elements such as lanes, boundaries, crosswalks, and traffic controls; semantic scene understanding; occupancy and free-space representations, including uncertain and occluded regions. - Develop scalable multimodal fusion architectures, including Transformer-based early, late, and hierarchical fusion; BEV, point/voxel, and image-based representations; temporal context aggregation; and cross-modal attention. - Create training, fine-tuning, and evaluation pipelines for large-scale multimodal datasets. - Define multi-task objectives and metrics that balance perception quality, geometric consistency, prediction accuracy, latency, and safety-critical behavior. - Investigate foundation-model approaches for autonomous driving, including vision-language models, multimodal pre-training, representation learning, and efficient deployment of learned world models. - Work closely with perception, mapping, prediction, planning, simulation, data, and embedded-software teams to convert research advances into robust, production-quality AV systems. - Develop systematic analysis and debugging tools for model failures, cross-sensor disagreement, long-tail scenarios, distribution shift, and regressions in closed-loop simulation and on-road evaluation. Check for any missing: The original had "including;" after "including;" maybe a typo; we handled. **Requirements / Qualifications** (What We Need To See). The paragraph: "What We Need To See BS, MS, or PhD in Computer Science, Robotics, Electrical Engineering, Machine Learning, or a related technical field, or equivalent experience.8+ years of experience, with at least 2+ years in the AV or robotics industry and 2+ years of leadership experience in a technically areaStrong experience developing production-quality sensor-fusion, perception, state-estimation, or autonomous-driving systems.Demonstrated experience with learning-based multimodal perception or fusion involving two or more cameras, LiDAR, radar, map, navigation, and ego-motion signals.Solid understanding of 3D geometry, coordinate frames, calibration, temporal synchronization, ego-motion compensation, tracking, uncertainty estimation, and sensor failure modes.Experience with deep-learning methods for 3D perception, lego-context scene representation, occupancy/occlusion prediction, semantic segmentation, object detection/tracking, motion prediction, or planning.Strong C++ and Python programming skills, with hands-on experience developing, training, and optimizing deep-learning models in PyTorch. Experience with CUDA