Company Description

Tondo Smart focuses on developing AI-powered IoT platforms designed to optimize, protect, and manage public infrastructure and urban airspace. By integrating advanced technology, Tondo Smart enables smarter, safer, and more sustainable communities. The company is dedicated to leveraging innovation to create intelligent solutions for urban challenges. Their work contributes to enhancing the quality of life in cities worldwide.

We are seeking a highly skilled Machine Learning Engineer with deep expertise in acoustic signal modeling/ Classification modeling to lead the development of advanced detection systems based on audio data.This role requires hands-on experience in training, optimizing, and deploying deep learning models for real-world acoustic environments, with strong production engineering capabilities on Linux systems.

• Role Responsibilities

·       Design and implement end-to-end ML pipelines for acoustic detection

·       Develop and optimize models using time-frequency representations (STFT, Mel spectrograms, etc.)

·       Design robust evaluation frameworks (ROC, PR curves, FAR/FRR, detection latency)

·       Engineer production-ready Python systems running on Linux

·       Build scalable, maintainable inference services and pipelines

·       Optimize models for real-time and edge deployment (e.g., Jetson, ARM-based systems)

·       Collaborate with acoustic experts and system engineers to refine detection performance

• Required Qualifications

·       3+ years of hands-on experience in ML model development using Python

·       Strong experience with PyTorch or TensorFlow

·       Proven experience in audio/acoustic ML (environmental sound classification, event detection, etc.)

·       Solid understanding of Digital Signal Processing (STFT, windowing tradeoffs, time–frequency resolution, filtering, sampling theory)

·       Experience handling imbalanced datasets and rare-event detection

·       Strong experience writing production-ready Python code on Linux systems (process management, concurrency, memory optimization, logging, packaging, deployment)

·       Experience building ML systems that run continuously in real-time environments

·       Proficiency with NumPy, SciPy, librosa, torchaudio, or similar

• Strong Advantage

·       Multi-microphone array processing

·       Real-time inference optimization (ONNX, TensorRT, quantization)

·       Edge deployment experience (Jetson, embedded Linux, ARM)

·       Experience with containerization (Docker) in Linux production environments

·       Background in surveillance, aerospace, or drone-related systems

• What We're Looking For

·       Deep technical ownership

·       Strong debugging skills in real-world noisy systems

·       Research mindset combined with production engineering discipline

·       Ability to bridge experimental ML and reliable operational systems