6 hours ago
Seoul, Korea, SouthSenior
Responsibilities
- Bring up and maintain the Fastlane Edge runtime on Sonatus and customer automotive control units.
- Integrate CAN/CAN-FD, DBC/ARXML, sandbox, Python runtime, security policies, cloud connectivity, and the Sonatus Southbound SDK.
- Package, deploy, validate, and optimize AI models using the Sonatus ADK, including concurrency, resource-budget, telemetry, and OTA verification.
- Port Python-developed models to native C++ and ONNX Runtime bundles with automated numerical-parity validation.
- Deliver Korean-language SDK and ADK training, office hours, integration documentation, and customer technical support.
- Run MLOps workflows for vehicle data ingest, alignment, training, model comparison, ONNX export, quantization, and on-target latency measurement.
- Tune virtual-sensor and estimation models on customer drives and manage ground-truth capture and versioned datasets.
- Execute vehicle validation drives, demos, acceptance testing, ASPICE evidence creation, on-site troubleshooting, and defect triage.
- Capture customer AI workload requirements and contribute fixes and hardening upstream to the toolchain.
- Use Claude and GitHub Copilot for recipe authoring, test generation, log analysis, and documentation within data-governance constraints.
Requirements
- Bachelor’s degree in Computer Science, Electrical Engineering, Mechanical/Control Engineering, or a related field; a master’s degree is preferred.
- 5+ years of experience in machine learning engineering or embedded software, including 2+ years deploying ML models on edge or embedded Linux targets.
- Strong Python skills and the ability to read, build, and debug modern C++14/17.
- Solid Linux debugging skills involving systemd, processes, logs, containers, and preferably cgroups, sandboxing, seccomp, or shared-library/glibc issues.
- Hands-on experience with structured device or bus-level signal data, including signal mapping, sampling, and time alignment.
- ML deployment experience with ONNX/ONNX Runtime and PyTorch or scikit-learn/XGBoost, with exposure to quantization and ARM profiling.
- Applied modeling experience with time-series or vehicle-signal data, including regression, estimation, or anomaly detection.
- Experience with reproducible ML pipelines, data tooling, experiment tracking, and versioned datasets.
- Python test automation, evidence-based triage, and willingness to divide time between development and validation.
- Fluent Korean, professional English, clear technical writing, and comfort supporting customer engineers and on-site test events.
- Preferred experience includes NXP S32/eIQ Auto, Qualcomm SA8xxx/SNPE-QNN, vehicle dynamics, virtual sensors, OEM IT and security stacks, ASPICE v4, ISO 26262, ISO 21448, LLM-assisted engineering, Jira, Confluence, Git-based workflows, and Agile/Scrum.
Benefits
- Hybrid office work arrangement at the Pangyo location.
- Point Reward Program for the Customer Integration team.
- Support for four social insurance plans.
- Maternity and paternity family leave.
- Complimentary lunches, snacks, and beverages during on-site working days.
- Wellness benefit allowance, phone and internet reimbursement, and computer accessory allowance.
- English support class and annual health checkup.
- Holiday appreciation gift.
Tech Stack
Categories
About Sonatus
Sonatus builds cloud-to-edge software for software-defined vehicles, including its Fastlane Platform that helps automotive OEMs and suppliers manage data, networking, and in-vehicle services with AI. It licenses its platform and tools to carmakers and Tier‑1s, and its technology is deployed in more than 8 million vehicles. Founded in 2018 and headquartered in Sunnyvale, California, the company is privately held with offices across North America, Europe, and Asia.