ML Engineer, Model Training - Senior Member of Technical Staff (SMTS)
⚲ Gdansk
Do uzgodnienia
Wymagania
- TensorFlow
- Python
Opis stanowiska
AMD is building a hardware-assisted security platform that uses silicon-level Performance Monitoring Counters (PMCs) and on-chip machine learning to detect advanced endpoint threats (ransomware, fileless malware, cryptojacking) at the processor layer, below OS-based evasion. The platform collects CPU behavioral telemetry, classifies it via an ML inference engine, and exposes threat signals to security-software partners through a standardized API. The team covers the full stack: silicon telemetry, ML training/validation, real-time inference, lab qualification and CI/CD.
Define and lead the ML model roadmap, progressing from binary malware/benign classification through multi-class threat taxonomy to behavioral attack-pattern detection.
Architect training pipelines that scale to a growing malware variant library with reproducible, versioned experiments; establish model quality gates for production promotion.
Lead research into advanced detection techniques including behavioral sequence modeling and detection of novel, previously unseen threat categories.
Optimize multi-class ML classifiers for NPU inference against throughput and latency requirements; collaborate with hardware teams on NPU capability requirements.
Drive dataset strategy including coverage across threat categories, synthetic data generation and dataset quality standards.
Mentor MTS ML engineers; lead model and code reviews; establish best practices for reproducibility, documentation and experimental rigor. Represent ML model strategy in architecture reviews, external partner technical meetings and potential research publications.
Define and lead the ML model roadmap, progressing from binary malware/benign classification through multi-class threat taxonomy to behavioral attack-pattern detection.
Architect training pipelines that scale to a growing malware variant library with reproducible, versioned experiments; establish model quality gates for production promotion.
Lead research into advanced detection techniques including behavioral sequence modeling and detection of novel, previously unseen threat categories.
Optimize multi-class ML classifiers for NPU inference against throughput and latency requirements; collaborate with hardware teams on NPU capability requirements.
Drive dataset strategy including coverage across threat categories, synthetic data generation and dataset quality standards.
Mentor MTS ML engineers; lead model and code reviews; establish best practices for reproducibility, documentation and experimental rigor. Represent ML model strategy in architecture reviews, external partner technical meetings and potential research publications.
🔍 Dekoder Ogłoszenia
🔴
Define and lead the ML model roadmap
Oczekuje się, że będziesz samodzielnie tworzyć i realizować strategię rozwoju modeli uczenia maszynowego, co może oznaczać dużą odpowiedzialność i konieczność podejmowania kluczowych decyzji technicznych.
🔴
progressing from binary malware/benign classification through multi-class threat taxonomy to behavioral attack-pattern detection
Projekt jest ambitny i obejmuje ewolucję od prostych zadań klasyfikacji do bardzo złożonych analiz behawioralnych, co może wymagać ciągłego uczenia się i adaptacji.
🔴
Architect training pipelines that scale to a growing malware variant library with reproducible, versioned experiments
Budowanie skalowalnych i powtarzalnych procesów uczenia maszynowego dla szybko rosnącej liczby wariantów złośliwego oprogramowania jest wyzwaniem technicznym, które może wymagać zaawansowanych umiejętności inżynierskich.
🟡
Mentor MTS ML engineers
Poza własnymi zadaniami technicznymi, będziesz odpowiedzialny za wspieranie rozwoju młodszych inżynierów ML, co wymaga umiejętności przywódczych i komunikacyjnych.
🟡
collaborate with hardware teams on NPU capability requirements
Wymaga ścisłej współpracy z zespołami sprzętowymi, co może oznaczać konieczność zrozumienia i komunikowania się w terminologii sprzętowej oraz wpływania na rozwój przyszłych technologii.