Job Details

International IT product company seeks a Software Engineer ML with 2+ years of experience in production ML, focusing on Speech & Audio. Full-time, hybrid role starting in Tashkent, then fully remote. Salary $3500-$5000.

We are a product team building an intelligent cloud telephony ecosystem for the US and Canadian markets. Our product is a fault-tolerant platform handling millions of traffic. ML is a core foundation of the system, operating in real-time. We are looking for an engineer with a deep understanding of audio model architecture to ensure their performance in a high-load production environment. Responsibilities: - Developing the AMD (Answering Machine Detection) system: retraining and tuning models for real-time call classification (distinguishing humans from answering machines/IVRs). - Full-cycle development: from collecting and 'dirty' labeling audio data to deployment and threshold calibration in production. - Integration into the Core product: migrating ML components into the backend infrastructure (C# / SIP / RTP stack) via ONNX Runtime. - Optimizing latency: striving for milliseconds in audio streaming conditions. - Deep Analysis: identifying errors and analyzing complex edge cases in real call scenarios. - Research (R&D): experimenting with noise reduction, VAD, and new speech processing architectures.

2+ years of experience in production ML (where your model actually worked with users). Practical experience with Speech/Audio: understanding of audio features and modern sound processing architectures. Engineering approach (QA-mindset): genuine interest in digging into data anomalies and stress-testing the system. Understanding of classic and modern concepts: Fine-tuning, Transfer Learning, and proficiency with metrics (Precision/Recall, ROC-AUC, Calibration). Ability to work end-to-end: from raw files to optimized inference. Key qualities: - Engineering autonomy: we value individuals who independently find problems and bring solutions to production. - Background: we highly welcome candidates coming to ML from Backend or QA, as code and testing culture are important to us. - Adaptability: the project is growing, tasks are numerous, and they directly impact the business. Plus: - Experience in Speech/Audio domain (ASR, VAD, Audio Classification). - Understanding of VoIP specifics and data streaming. - Experience with MLOps and model monitoring tools.

Mandatory offline onboarding in Tashkent (2-3 months) for product immersion, followed by full remote work. Real production tasks in an international, high-load product. Opportunities for professional growth and compensation review as tasks become more complex. Work in a team with strong engineering expertise and no bureaucracy.