DSML.KZ Community Hub

AI platform for communication. Responsibilities: • Develop, fine-tune, evaluate, and implement generative models for image, video, text, and multimodal applications. • Build scalable data, training, evaluation, real-time inference, and MLOps pipelines; optimize latency, throughput, memory usage, quality, and infrastructure costs. • Integrate models into production systems, automate deployment and monitoring, and research and assess generative AI methods for production use. Requirements: • 3+ years of professional machine learning and software development experience; practical experience developing or deploying generative AI models; deep understanding of diffusion models and modern image or video-generation architectures; experience with LLM prompt engineering, fine-tuning, evaluation, structured outputs, tool invocation, and behavior management; PyTorch or TensorFlow; productionizing models and operating reliable inference services; deployment on GPU servers and cloud platforms;- Optional: • Experience with preference optimization methods such as RLHF, DPO, or reinforcement learning; constrained decoding, function calling, agent systems, safety mechanisms, red-team testing, and automated LLM evaluation; multimodal models; video-generation quality improvements; GPU optimization; MLOps platforms, model registries, experiment tracking, observability, CI/CD, and infrastructure as code.