---
title: "Senior 3D Perception & Sensor Fusion ML Engineer"
slug: "senior-3d-perception-sensor-fusion-ml-engineer"
id: "174c4fdb-559f-4211-a66e-1bbb44c57c2e"
department: "Robotics & Embodied AI"
department_code: "ROB"
location: "Oxford, UK"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "£85k – £120k (annual package)"
is_remote: false
is_featured: false
headcount: 1
canonical_url: "https://jobs.talent247.co/jobs/senior-3d-perception-sensor-fusion-ml-engineer"
markdown_url: "https://jobs.talent247.co/jobs/senior-3d-perception-sensor-fusion-ml-engineer.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-3d-perception-sensor-fusion-ml-engineer/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-3d-perception-sensor-fusion-ml-engineer/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:42:57.583Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Senior 3D Perception & Sensor Fusion ML Engineer

## Position Telemetry

- **Reference:** `SENIOR-3D-PERCEPTI`
- **Practice / Sector:** Robotics & Embodied AI (`ROB`)
- **Location:** Oxford, UK
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** £85k – £120k (annual package)
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/senior-3d-perception-sensor-fusion-ml-engineer

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is an Oxford-headquartered autonomous vehicle software leader commercializing universal, safety-certified autonomy across industrial logistics yards, airports, ports, mines, and commercial passenger shuttles. Unlike fair-weather highway demos, their fleets operate 24/7 in complex, dynamic industrial environments alongside heavy machinery and pedestrians under heavy rain, fog, dust, and night-time conditions.

They have partnered with Talent247 to hire a Senior 3D Perception & Sensor Fusion ML Engineer in Oxford. Joining their Core Perception squad, the appointed engineer will design, train, and deploy multi-modal 3D bird's-eye-view (BEV) detection, 4D occupancy, and multi-object tracking networks fusing camera, LiDAR, and imaging-radar streams onto onboard vehicle compute.

KEY RESPONSIBILITIES
• Architect and train spatiotemporal multi-sensor fusion models (BEV transformers, 4D semantic occupancy networks, point-cloud + image fusion) delivering centimetre-accurate 3D understanding in unstructured and adverse-weather environments.
• Design self-supervised pre-training, auto-labeling teacher distillation, and active-learning loops over terabytes of multi-fleet driving logs to slash manual 3D cuboid annotation costs.
• Optimize onboard model architectures via quantization-aware training (INT8/FP16), structured pruning, and TensorRT/C++ graph compilation to hit strict real-time latency and determinism budgets on vehicle GPUs.
• Build granular safety-critical evaluation slices targeting occluded pedestrians, degraded sensor dropouts, calibration perturbations, and rare industrial obstacles.
• Collaborate closely with sensor-calibration, motion-planning, and onboard systems engineers from lab hardware-in-the-loop rigs through live test-track and customer fleet rollouts.
• Mentor fellow perception engineers and contribute to architectural roadmaps across next-generation sensor suites.

TECHNICAL ENVIRONMENT & OFFERING
Their perception stack runs on Python, PyTorch, C++17/20, CUDA/TensorRT, ROS2/custom deterministic middleware, and cloud GPU training clusters. Based hybrid in Oxford, our client offers £85,000–£120,000 GBP base salary plus equity options, generous UK pension, 28 days holiday + bank holidays, private healthcare, and UK Skilled Worker visa sponsorship.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 5+ years of industry experience in 3D computer vision and deep learning perception for autonomous vehicles, mobile robotics, or embodied AI.
• Deep hands-on proficiency in PyTorch for training 3D object detection, segmentation, occupancy, or multi-object tracking networks across LiDAR, camera, and/or radar modalities.
• Solid C++ and CUDA/TensorRT skills for optimizing and integrating trained perception models into real-time onboard vehicle pipelines.
• Strong geometric fundamentals in camera projection models, extrinsic/intrinsic sensor calibration, ego-motion compensation, and temporal fusion.
• Proven ability to diagnose real-world fleet failure modes and translate them into measurable model and dataset improvements.

PREFERRED / BONUS EXPERIENCE
• MSc or PhD in Robotics, Computer Vision, or Machine Learning, or publications at CVPR, ICCV, ECCV, ICRA, or IROS.
• Experience with foundation vision backbones, distillation from offboard 3D auto-labelers, or safety-critical ISO 26262 / SOTIF validation workflows.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/senior-3d-perception-sensor-fusion-ml-engineer or via HTTP `POST https://jobs.talent247.co/api/public/jobs/senior-3d-perception-sensor-fusion-ml-engineer/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/senior-3d-perception-sensor-fusion-ml-engineer/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
