ML Engineer

Requirements:

• Experience with PyTorch and model optimization for production
• Compilation of ML models into native code (ONNX, TorchScript, experience with model quantization)
• Working with GPU (CUDA, inference optimization on graphics accelerators)
• Experience deploying models in a closed circuit without Internet access
• Containerization of ML solutions (Docker, understanding security requirements)
• Experience with quantized models for edge computing
• Understanding of time series and streaming data
• Experience in classifying network traffic or real-time tasks
• Working with large volumes of data (terabits of traffic)

Technical Stack:

• Python (with the ability to compile via Cython)
• C/C++ for low-level optimization
• LangGraph/Pydantic AI for agent systems
• Vector databases (Weaviate, Qdrant)
• Ollama for local LLM deployment
• Hadoop, Hive, Spark for big data processing
• Kafka for streaming data

Additional Requirements:

• Readiness to work with closed systems without external dependencies
• Understanding of security requirements in the telecom industry
• Experience in international projects (the system is deployed in different countries)
• Knowledge of English (documentation, communication with DPI library vendors)
• Experience in managing data science projects
• Networking concepts (TCP/IP, RADIUS, GTP)
• Linux system configuration
• Git/GitHub for version control

A plus:

• Experience with DPI systems (Sandvine, Procera, Allot)
• Familiarity with DPDK and high-performance packet processing
• Experience with Rohde & Schwarz or Enea Qosmos SDK
• Understanding of network protocols and the structure of Internet traffic
• Experience with Dynamic Content Classification
• Working with Quality of Experience (QoE) models
• Network Intelligence and high-precision traffic classification

Education & Experience:

• BE/ME in Computer Science or Engineering
• 2+ years of practical experience in data science/ML