Free with a WorkJio account: we compare your profile with this job and show what you already meet.
Share with a friend
Start
Not stated
Work mode
On-site
Experience
Some experience
About the role
You will join the Robotics Technology team, which builds urban embodied AI for robot delivery services. The team designs hardware in-house, develops control and machine-learning systems, and tests them in real-world conditions and production fleet operations. This role suits an experienced perception engineer who wants to take research from prototype to fleet deployment and directly influence last-mile logistics.
What you'll do
Develop and improve multi-object tracking, covering data association, Bayesian state estimation, track lifecycle and ID stability through occlusions and crowded scenes, and move toward learning-based tracking.
Build generic and open-set world understanding, including class-agnostic obstacle detection, occupancy grids and networks, clustering and learned generic-object branches.
Build motion prediction for pedestrians, cyclists, vehicles and other agents, covering multi-modal trajectory forecasting, interaction-aware prediction, occupancy flow and uncertainty estimation.
Develop and evaluate end-to-end temporal perception-prediction approaches such as joint detection-tracking-forecasting, streaming BEV representations and learned world models, while keeping modular baselines and production fallbacks.
Explore VLM, VLA and embodied foundation models for open-vocabulary perception, semantic scene reasoning, long-tail discovery and auto-labeling, then distil or adapt useful capabilities for on-robot deployment.
Strengthen multi-sensor and temporal integration across camera, LiDAR, radar and IMU so geometry, timing, identities, predictions and semantic context stay consistent over time.
Define quality gates across tracking, prediction, occupancy recall, closed-loop safety outcomes and embedded latency budgets, and drive down regressions with fleet-log root-cause analysis.
Partner with Detection, Prediction, Planning, Data & Infra, and Integration teams to move capabilities from research into reproducible training, replay, simulation, NVIDIA Orin / TensorRT deployment and production-like robot testing.
Good to know
You will report to the Senior Principal Perception & Prediction Engineer.
The role is onsite at a Grab office in Singapore.
The team is based in Singapore and China.
Requirements
At least 3 years hands-on experience shipping or operating perception and/or prediction systems for robotics, AV or ADAS, with strong autonomous-driving experience.
Experience across classical/geometric and learning-based approaches, with judgement on modular, end-to-end or hybrid designs and interpretable safety fallbacks.
Proven depth in multi-object tracking: data association, Bayesian filtering, track lifecycle, multi-sensor/temporal association, ID stability and low-latency evaluation.
Hands-on experience with generic/open-set perception: class-agnostic obstacle detection, occupancy representations, LiDAR clustering, anomaly handling or fusion with learned detections.
Hands-on depth in at least one modern temporal area: motion forecasting, occupancy flow, streaming/temporal BEV, learned world models, or joint detection-tracking-prediction.
Fundamentals in geometry and multi-sensor systems, including practical calibration and diagnosing alignment or time-synchronisation issues.
Jio can write this application for you
Jio reads this job and your profile, then writes a short letter in your words. You read it, change anything, and send it yourself. Nothing is sent until you say so.
