Doktorand*in Embedded AI Systems: Laufzeitoptimierung für Transformer-Modelle

Tasks

• Analyze reinforcement learning policies versus static heuristics
• Develop adaptive scheduling algorithms for edge runtime
• Evaluate transformer inference on embedded hardware
• Formalize runtime orchestration as optimization problem
• Implement performance critical runtime components in C or C++
• Integrate runtime system into overall architecture
• Measure latency throughput and energy efficiency
• Prototype and simulate in Python
• Publish research at international conferences and journals
• Supervise bachelor and master theses

Perks/Benefits

• Corporate benefits
• Flexible working hours
• Job ticket subsidy
• Mobile working
• Pension plan

Skills/Tech-stack

ARM Cortex | ARM Cortex A | ARM Cortex-M | C# | C++ | Cortex A | Cortex-M | FPGA | HuggingFace Transformers | JAX | Machine Learning | ONNX | Online Optimization | Optimization | PyTorch | Python | RISC-V | Real Time | Real-time Systems | Reinforcement Learning | Resource allocation | Scheduling algorithms | Time Systems | Transformer