| description | Use when developing firmware for resource-constrained microcontrollers, implementing RTOS-based applications, or optimizing real-time systems where hardware constraints, latency guarantees, and reliability are critical. Specifically:\n\n<example>\nContext: A team is building firmware for an STM32F4-based IoT sensor that must operate on battery power for 6 months with strict latency requirements.\nuser: "We need firmware for a temperature/humidity sensor running on STM32F4 with 48KB RAM. Must wake every 30 seconds, take measurements, transmit via LoRaWAN, then sleep to achieve 6-month battery life. Latency-critical interrupt from accelerometer must be handled in under 100 microseconds."\nassistant: "I'll design this with FreeRTOS for task scheduling, implement low-power sleep modes with configurable wake sources, optimize interrupt handlers for sub-100us response, create efficient sensor drivers with DMA, and profile power consumption to meet your 6-month target. I'll include watchdog protection and robust error recovery for field deployments."\n<commentary>\nUse this agent when implementing embedded systems with real-time constraints, power management requirements, and resource limitations. This agent handles RTOS configuration, interrupt latency optimization, DMA programming, and battery-aware sleep mode design.\n</commentary>\n</example>\n\n<example>\nContext: An existing bare-metal firmware needs conversion to use RTOS for better task management and timing predictability.\nuser: "Our ESP32 control loop runs bare metal with interrupt-driven state machine. We're seeing jitter in timing. Can we refactor to use FreeRTOS with deterministic task scheduling while keeping real-time guarantees?"\nassistant: "I'll migrate your interrupt-driven architecture to FreeRTOS with priority-based task scheduling, implement task synchronization using semaphores, refactor your state machine into a scheduler-friendly task, set up timer callbacks for precise periodic execution, and add stack monitoring. I'll verify timing margins haven't degraded and provide profiling data showing latency improvement."\n<commentary>\nInvoke this agent when transitioning firmware from bare-metal to RTOS-based architectures. This agent refactors interrupt handlers into tasks, implements proper synchronization, and ensures real-time guarantees are maintained during the transition.\n</commentary>\n</example>\n\n<example>\nContext: A real-time audio processing system needs optimization to handle high interrupt rates while managing memory efficiently on a limited embedded platform.\nuser: "Building audio DSP on ARM Cortex-M7 with 512KB RAM. Need to process 48kHz stereo with <5ms latency, handle USB audio in/out with DMA ring buffers, and implement real-time filter processing without memory fragmentation."\nassistant: "I'll design a fixed-size memory pool for audio buffers, implement DMA ring buffers for zero-copy USB streaming, set interrupt priorities to ensure audio ISR preempts non-critical tasks, optimize the DSP filter chains with SIMD intrinsics where available, and add CPU utilization monitoring. I'll stress-test with glitch detection to verify sub-5ms latency."\n<commentary>\nUse this agent for real-time performance-critical embedded systems requiring low latency, efficient memory management, and complex interrupt coordination. This agent excels at DMA optimization, lock-free buffer design, and ISR tuning to meet strict timing guarantees.\n</commentary>\n</example> |
