Embedded Machine Learning Engineer (AI/ML)

Responsibilites

• Prototype Development: Lead rapid prototyping of ML models for edge intelligence across Voice, Sense, and Control domains, tightly integrated with Cirrus Logic’s mixed-signal processing strengths.
• Data & Model Engineering: Build datasets, design model architectures, and optimize performance, efficiency, and interpretability. Explore advanced approaches in ML-augmented signal processing, anomaly detection, and adaptive control.
• System Integration: Collaborate with silicon, firmware, and systems teams to co-design ML architectures that operate efficiently on constrained hardware and embedded systems, balancing algorithmic accuracy with compute and power budgets.
• Exploration & Research: Stay at the forefront of ML frameworks, foundation/SLM trends, and physical-world AI applications. Scout external IP, academic work, and startups to inform CVL’s ML strategy.
• Mentorship & Technical Leadership: Provide guidance and technical direction to away-team engineers and contributors across Cirrus Logic. Share best practices in ML model lifecycle, from experimentation to deployment.
• Cross-Functional Collaboration: Work hand-in-hand with Innovation Managers, advisory teams, customers,  and external partners to identify opportunities, define success criteria, and validate ML-enabled innovations in real-world scenarios.
• Impact Assessment: Help define benchmarks, evaluation metrics, and pass/fail criteria that ensure ML prototypes address significant industry problems with clear paths to monetization.

Required Skills and Qualifications

• Educational Background: Master’s or Ph.D. in Computer Science, Electrical Engineering, or related field with a focus on ML/AI.
• Experience: 8+ years of hands-on experience developing and deploying ML systems on the Edge and within embedded platforms, including ownership of datasets, model development, and deployment pipelines. Proven experience implementing ML inference on resource-constrained systems such as microcontrollers, embedded SoCs, or custom silicon.
• Architectural Expertise: Demonstrated experience with CNNs, RNNs (LSTM/GRU), and Transformer-based models, including custom architecture design and optimization for production. Experience tailoring these architectures for low-latency and low-power embedded inference.
• Technical Depth: Strong understanding of representation learning, attention mechanisms, sequence-to-sequence modeling, and generative architectures. Ability to translate these methods into efficient implementations suited for real-time sensor, audio, or control workloads.
• Optimization for Edge: Experience with quantization, pruning, knowledge distillation, mixed-precision training, and compiler-level optimizations to deploy models on CPUs, DSPs, NPUs, or hybrid SoC architectures. Familiarity with memory hierarchy tradeoffs, compute-offload, and bandwidth constraints in embedded ML.
• Embedded & Firmware Integration: Proficiency in embedded software and firmware development (C/C++/Python) with experience integrating ML inference engines into real-time embedded stacks, RTOS environments, or bare-metal systems. Understanding of firmware pipelines, peripheral I/O, and signal-path integration for ML-augmented mixed-signal systems.
• Data Engineering: Ability to design labeling strategies, synthetic data generation, and augmentation pipelines to support robust model development. Understanding of data acquisition and preprocessing directly from embedded sensors.
• Systems Thinking: Proven track record of co-designing ML and firmware solutions alongside hardware teams, balancing algorithmic, architectural, and physical constraints. Familiarity with embedded ML frameworks and toolchains (e.g., TensorRT, ONNX Runtime, TVM, CoreML, TFLite, Glow, Edge Impulse).
• Collaboration & Communication: Ability to translate complex ML concepts into actionable insights for cross-disciplinary teams of algorithm, firmware, and hardware engineers.

Preferred Skills and Qualifications

• Startup & Incubator Experience: Background in early-stage, high-ambiguity environments; experience contributing to incubation of new products or platforms.
• Specialized ML Expertise: Experience in one or more of: generative models for voice, time-series/sequence modeling, anomaly detection for sensors, reinforcement learning for control systems.
• Tooling: Familiarity with MLOps frameworks, data labeling pipelines, and distributed training.
• External Engagement: Experience collaborating with startups, academic labs, or open-source communities.
• Business Acumen: Ability to assess the business and monetization value of ML solutions in emerging markets.