We use PostHog and Clarity to improve the site, and Meta Pixel to measure page views and ad effectiveness. Usage data may include page addresses and browser identifiers. You can opt out of collection at any time. Details in the Privacy Policy
Low-Level Software Engineer - JB-765 at PwC | NEXT Technology Solutions · Trevo
Low-Level Software Engineer – EDGE Team
Develop and implement a core software platform that enables AI model execution across various EDGE hardware devices. This role involves low-level software development, seamless hardware integration (CPU, GPU, and AI accelerators), system resource management, and performance optimization to ensure efficient, stable, and low-latency local inference under strict resource constraints.
What you'll do
EDGE Application Development: Build a central C/C++ application to manage, load, and run local AI model inference on edge devices, designing unified interfaces that abstract hardware differences.
Hardware Porting & Adaptation: Adapt the software runtime and environment to diverse platforms (e.g., NVIDIA Jetson, SIMA), working closely with APIs, drivers, and hardware components while respecting device limitations.
Low-Level System Development: Write system-level software in Embedded Linux environments, handle memory management, implement multithreading/concurrency, and debug complex hardware-software interfaces.
Performance & Optimization: Conduct profiling and benchmarking to identify bottlenecks, optimizing CPU, GPU, memory, and I/O usage to improve latency, throughput, and power efficiency.
Integration & Lab Testing: Integrate AI models onto real hardware, perform rigorous integration and stability tests in the lab, and document setup, optimization, and testing procedures.
Requirements
Experience: 3–5+ years of software development experience, with a strong focus on Low-Level, Embedded, or System Software.
Languages & OS: Advanced proficiency in C and C++, alongside significant hands-on experience with Linux.
System Architecture: Solid understanding of computer architecture, memory management, multithreading, concurrency, and hardware-software interactions.
Performance Optimization: Proven experience in performance tuning, using profiling tools, benchmarking, and developing for low-latency, resource-constrained environments.
Key Traits: Strong hands-on engineering capabilities, deep debugging skills for complex system issues, a systemic mindset (understanding the Model → Runtime → Software → Hardware pipeline), and the ability to learn new platforms independently.
Nice to have
Hands-on experience with NVIDIA Jetson and/or SIMA platforms.
Experience with AI/ML inference at the Edge, TensorRT, and various AI accelerators/runtimes.
Proficiency in CUDA / GPU programming or ARM architecture.
Background in Embedded Linux, Drivers, BSP, Kernel development, and Cross Compilation.