Embedded Firmware Principal Architect

We are seeking an exceptional Embedded Firmware Principal Architect to serve as the top
individual-contributor technical authority for our organization. This role demands a rare
combination of technical depth, architectural vision, and AI/ML knowledge to define the technical
direction for embedded software and edge intelligence systems, working hands-on and in close
partnership with embedded software developers, validation engineers, and data scientists
delivering mission-critical embedded and edge intelligence software solutions for Ambient
Scientific AI SoCs.
This is a high-impact individual-contributor position for a technologist who thrives at the
intersection of embedded engineering excellence and edge AI innovation. You will set the
technical direction, drive architectural decisions, guide the development of data science models
deployed on resource-constrained edge hardware, and own deep hands-on delivery of
embedded software across the full development lifecycle — influencing engineering teams
through technical authority and mentorship rather than formal management.
💻 Embedded Software
✅ Validation & Quality
🧠 Data Science
Firmware, BSP, drivers, RTOS,
middleware, application software
SW validation, test automation,
quality assurance
ML/AI model development, edge
deployment, sensor fusion, data
pipelines
Key Responsibilities
Technical Strategy & Direction
●Act as the top technical authority for embedded software and edge AI, owning the
technical roadmap and engineering vision in partnership with business objectives.
●Partner with executive leadership (VP Engineering, CTO, CEO) to shape the multi-year
software and data science technical strategy for the embedded systems division.
●Drive cross-functional technical alignment with Hardware, Systems Engineering, Product
Management, and Program Management teams.
●Build and communicate a unified technical vision where embedded software and edge
AI/ML capabilities are co-designed for optimal real-world performance.
●Champion the adoption of AI-assisted tools for quick, high-quality software development,
automated unit testing, and full software validation.
Software Architecture & Technical Governance
●Serve as the primary Software Architect, defining and governing embedded system
software architectures including BSP, middleware, RTOS, device drivers, and application
layers.
●Lead architecture design reviews, establish design patterns, coding standards, and
ensure architectural integrity across all software subsystems.
●Make critical technical decisions on platforms, toolchains, SDKs, and RTOS selection
(FreeRTOS, Zephyr).
●Define the edge AI/ML software architecture: model inference pipelines, sensor data
ingestion layers, pre/post-processing frameworks, and on-device model lifecycle
management.
●Champion software reusability, modularity, and scalability; drive adoption of model-based
design and component-based architectures where applicable.
●Own the software architecture documentation and ensure adherence to relevant
standards.
Hands-On Software Development
●Remain deeply engaged in software development — writing, reviewing, and debugging
embedded C/C++ and Python code as the primary technical contributor on critical path
items.
●Conduct deep technical reviews of system-level software: bootloaders, firmware update
mechanisms, memory management, interrupt handling, power management, and real-
time scheduling.
●Lead proof-of-concept development for new hardware bring-ups, new SoC/MCU/NPU
platforms, and emerging edge AI technology integrations.
●Personally drive root-cause analysis on high-severity software defects, model inference
failures, system-level integration issues, and field escalations.
Data Science & Edge AI Technical Leadership
●Provide technical leadership to the Data Science Engineering function on developing,
optimizing, and deploying ML/AI models targeting edge and embedded hardware.
●Define the end-to-end edge AI development workflow: data collection and labeling
pipelines, model training infrastructure, quantization and pruning strategies, deployment
packaging, and on-device validation.
●Guide data scientists in selecting appropriate model architectures (CNNs, RNNs,
Transformers, GNNs, classical ML) optimized for inference on Ambient Scientific existing
and upcoming SoCs.
●Oversee model compression techniques — quantization (INT8/INT4), pruning,
knowledge distillation, and neural architecture search — to meet embedded memory,
latency, and power constraints.
●Champion edge inference frameworks (like TensorFlow Lite, STM32Cube.AI) for
Ambient Scientific edge AI SoCs.
●Drive sensor fusion model development integrating data from IMU, radar, camera,
temperature, acoustic, and other embedded sensor modalities.
●Establish model performance benchmarking standards: accuracy, latency, memory
footprint, energy consumption, and robustness under real-world operating conditions.
●Ensure data science work is tightly integrated with the embedded software team — co-
designing data interfaces, inference APIs, and real-time triggering mechanisms.
Technical Mentorship & Cross-Disciplinary Influence
●Mentor senior, staff, and principal engineers and data scientists through design reviews,
architecture discussions, and hands-on pairing — influencing through technical authority
rather than formal reporting lines.
●Bridge the cultural and technical gap between embedded engineers and data scientists.
●Act as a technical role model, raising the engineering bar through example, code review,
and design leadership.
Software Validation & Quality Engineering
●Define embedded software validation strategy: test plans, coverage requirements, and
validation methodologies (unit, integration, HIL, SIL, regression).
●Extend validation practices to cover ML model validation: dataset quality audits, model
robustness testing, edge-case coverage, adversarial testing, and on-device accuracy
benchmarking.
●Champion shift-left testing practices, integrating validation early in the development
process.
●Drive adoption of automated testing frameworks, CI/CD pipelines for embedded targets
and ML models, and static analysis tooling.
●Champion quality gate enforcement: code coverage targets, static analysis compliance,
model performance thresholds, peer review completion, and defect escape rate metrics.
Process, Tools & Delivery
●Define and continuously improve development processes aligned with Agile across
embedded, validation, and data science workstreams.
●Own the software development environment: version control strategy (Git/DVC),
branching models, build systems (CMake, Make, Yocto), MLOps pipelines, and release
management.
●Drive technical risk management: proactively identify risks across firmware, AI model
performance, and data pipeline reliability; develop mitigation plans and communicate to
stakeholders.
●Report technical status, architectural milestones, and quality metrics to senior leadership
with clarity and transparency.
Required Qualifications
Education
●Bachelor's degree in Computer Engineering, Electrical Engineering, Computer Science,
Data Science, or related technical field.
●Master's degree or Ph.D. in a relevant field preferred.
Experience
●12+ years of progressive embedded software or software engineering experience, with
demonstrated experience operating at a principal/staff architect level.
●Demonstrated experience acting as a Software Architect or principal-level technical
authority, even if not in title.
●Proven hands-on technical contributor: ability to read, write, and review production-
quality embedded C/C++ code and ML/AI development artifacts.
●Experience technically guiding or closely collaborating with engineers across software
development, validation, and/or data science functions.
●At least 2 years of direct technical collaboration with a data science or ML engineering
team.
Embedded Software Technical Expertise
●Languages: Embedded C/C++, Python
●RTOS / OS: FreeRTOS, Zephyr, or equivalent production RTOS experience
●Processors: MCU/MPU/NPU architectures: ARM

Work arrangement
No