Summary
Andrew Stepin is an AI/ML software engineer and former product leader with a decade-plus background blending hands-on machine learning engineering, MLOps, and product management across banking, ecommerce and HR SaaS. He currently builds production-grade ML systems using PyTorch/TensorFlow/JAX, Transformers and a modern inference stack (ONNX, TensorRT, Triton, vLLM) while operating cloud-native MLOps on AWS/Azure/GCP, Kubernetes and Databricks. Previously he scaled marketplaces and HR platforms end-to-end as a product manager, driving 3x revenue growth, real-time KPI pipelines and A/B testing frameworks that measurably improved conversion and adoption. He also contributed to TensorFlow documentation at Google, mentoring international contributors and shipping localized tutorials—an unusual blend of deep technical authoring and engineering. Comfortable moving between C++, Python and TypeScript, Andrew pairs research-minded model work (GANs, generative projects) with pragmatic delivery of enterprise-grade systems. Based in Moscow, he’s fluent at turning experimental ML into robust, deployable products.
5 years of coding experience
MS, Experimental Nuclear Physics, Astrophysics, MS, Experimental Nuclear Physics, Astrophysics at National Research Nuclear University MEPhI (Moscow Engineering Physics Institute)