Utilidata is a fast-growing NVIDIA-backed edge AI company enabling greater visibility and control of power utilization in energy-intensive infrastructure, like the electric grid and data centers. Karman, the company’s distributed AI platform powered by a custom NVIDIA module, is transforming the way utility companies operate the grid edge and will enable data centers to unlock more compute for the same provisioned power.
Utilidata is seeking a Platform Architect to define and lead the architecture of our Karman data center solution, focused on high frequency waveform acquisition, low latency control paths, and the Karman SoM hardware and software co-design for GPU-dense environments. This role requires deep technical authority across embedded systems, operating systems, firmware, high performance software, and accelerator optimization, with a strong emphasis on deterministic latency and end to end power-aware system design. The Architect will work closely with engineering leadership, hardware, firmware, platform, and data teams to ensure Karman delivers actionable insight and control at data-center scale.
Responsibilities- Architecture & Technical Strategy
- Own the end-to-end architecture of Utilidata’s Karman data center platform, spanning:
- High-resolution waveform sensing and acquisition
- Embedded firmware and Yocto based OS layers
- Edge compute and control loops
- Off device visibility, analytics, and orchestration
- Define architectural principles that prioritize determinism, reliability, security, and low latency in power critical environments for production.
- Architect systems that integrate GPU telemetry, power states, and control interfaces.
- Serve as a technical leader and mentor for sub-system level principal engineers and architects across hardware, firmware, operating system layers, and on-device software teams.
- Partner with Product and Customer teams to align architecture with real world data center constraints.
- Drive technical reviews, architectural documentation, and long-term roadmap planning.
- Translate product and customer requirements into technical designs that meet near term business commitments while establishing a clear path to long term scalability.
- Product Development & Technical Leadership
- Architect ultra low latency software paths for waveform capture and processing, real time anomaly detection and closed loop power control
- Ensure system designs support predictable response times suitable for power capping and spike mitigation in GPU-dense server racks
- Lead performance modeling, instrumentation, and optimization across the full stack.
- Provide deep technical leadership in:
- Low-level OS internals, kernel scheduling, memory management, and I/O paths
- Firmware, bootloaders, BSPs, and hardware abstraction layers
- Real-time and near real time execution environments
- Partner with hardware teams on sensor integration, signal fidelity, timing accuracy, and compute placement
Minimum Qualifications - 10+ years experience in systems architecture, embedded systems, or high-performance platforms
- Prior experience designing and delivering distributed embedded systems from initial architecture through production build and large scale deployment in mission-critical environments
- Deep expertise in low level operating systems, firmware, and embedded software
- Proven track record designing low latency, high throughput, real time or near real time systems
- Strong background in hardware-software co-design, especially in power-constrained or performance critical environments
- Experience with performance profiling, tracing, and optimization across CPU, memory, networking, and I/O
- Experience with IoT or edge platforms that operate at scale and under strict reliability constraints
- Familiarity with real-time scheduling, timing analysis, and deterministic system behavior
- Proven ability to lead and influence across teams and organizational levels without formal authority by building trust and driving alignment
- Proven ability to communicate effectively across all organizational levels and translate complex information into clear, actionable guidance
- Demonstrated ability to lead architecture across multiple teams and disciplines
- Willingness to travel up to 20% of time
Enhanced Qualifications (Nice to Have) - Experience working with GPU based systems or other accelerators, including power and performance tradeoffs
- Familiar with NVIDIA SOC architectures and developing productionizable implementations
- Experience in data center infrastructure, power systems, or energy aware computing
- Familiarity with RDMA based architectures, high speed interconnects, and zero copy data pipelines
- Exposure to AI/ML inference pipelines, especially in real time or edge contexts
Salary Range: $195,000 to $235,000 base compensation depending on experience. Salary will be commensurate with an individual's skills, training, years of experience, and in line with internal compensation bands.
Location: This position can be performed remotely from anywhere in the United States (preference for candidates based in close proximity to the company's headquarters in Ann Arbor, MI)
Our Commitments:Utilidata values the diversity of our team. We provide equal employment opportunities without regard to race, color, religion, creed, sex, gender, sexual orientation, gender identity or expression, national origin, age, physical disability, mental disability, medical condition, pregnancy or childbirth, sexual orientation, genetics, genetic information, marital status, or status as a covered veteran or any other basis protected by applicable federal, state and local laws.
We are committed to:
- Creating a diverse and inclusive workplace that is welcoming, supportive, affirming and respectful
- Empowering employees to solve problems and work together to make a difference
- Providing mentorship and growth opportunities as part of a collaborative team
- A flexible work environment with flexible paid time off
- Competitive compensation and benefits, including health, dental, vision, and employer-match 401k
