SRE, AI Inference Engineer

At F5, we strive to bring a better digital world to life. Our teams empower organizations across the globe to create, secure, and run applications that enhance how we experience our evolving digital world. We are passionate about cybersecurity, from protecting consumers from fraud to enabling companies to focus on innovation.  Everything we do centers around people. That means we obsess over how to make the lives of our customers, and their customers, better. And it means we prioritize a diverse F5 community where each individual can thrive. Job Description The  AI Inference Engineer  plays a critical role in the AI lifecycle by bridging the gap between high-performance model development and optimized deployment environments. This position focuses on optimizing  Large Language Models (LLMs)  for inference, serving diverse environments—from GPU-rich data centers to resource-constrained edge devices—with a strong emphasis on maximizing throughput, minimizing latency, and maintaining model accuracy.   This role is pivotal in advancing F5’s AI capabilities, ensuring enterprise-grade reliability by leveraging hardware acceleration, designing scalable infrastructure, and monitoring system performance.  Key Responsibilities   High-Performance AI Serving   Build and maintain robust inference engines using tools like  vLLM ,  TGI (Text Generation Inference) , and  NVIDIA Triton , ensuring high performance at scale.   Handle deployment optimizations to deliver low-latency AI serving solutions for multiple business applications.  Hardware Acceleration and Optimization   Profile and optimize models for specialized hardware backends, including  NVIDIA GPUs  (CUDA/TensorRT),  Apple Silicon (CoreML) , and AI accelerators like  TPUs  and  LPUs .   Collaborate with hardware teams to maximize utilization and performance across various computational environments.  Inference Orchestration and Scalability   Design and implement  auto-scaling architectures  for online (real-time) and batch inference pipelines, leveraging  Kubernetes  for inference routing and orchestration.   Ensure software solutions are optimized for peak performance during traffic spikes, maintaining reliability and scalability.  Performance Monitoring and Observability   Establish robust observability frameworks to monitor  Time to First Token (TTFT) , tokens per second, and memory bandwidth utilization against service-level agreements (SLAs).   Build and execute  performance and load testing suites  to identify bottlenecks and ensure consistent reliability at scale.  Technical Requirements   Required Skills:   Programming Languages:  Proficiency in programming languages such as  Python ,  C++ ,  Rust , or  Golang  specifically for high-performance AI workflows.   Inference Tools:  Proven hands-on experience with tools like  vLLM ,  TensorRT ,  Llama.cpp , and  Ollama  for inference development and optimization.   Infrastructure Expertise:  Strong familiarity with infrastructure technologies, including  Docker ,  Kubernetes , and cloud platforms such as  AWS ,  GCP , and  Azure .   Hardware Optimization Expertise:  Comprehensive understanding of GPU and AI hardware, including techniques for profiling and optimizing performance for accelerators like NVIDIA GPUs and TPUs.  Preferred Experience:   Prior experience deploying  Large Language Models (LLMs)  with advanced techniques like  Speculative Decoding  or  PagedAttention .   Contributions to  open-source inference libraries  or hardware-level kernel development (e.g., CUDA, Triton kernels).   Background in  MLOps  or  SRE  roles focused on  high-performance AI endpoints  and reliability during demand surges.   Proficiency in designing scalable solutions for high-throughput inference environments optimized for traffic bursts.  Success Metrics (KPIs):   Latency Reduction:  Continuously improve inference latency metrics, ensuring minimal  Time to First Token (TTFT)  and maximum tokens per second.   Cost Efficiency:  Achieve lower "Cost per 1K Tokens" through better resource utilization and hardware optimization.   Scalability:  Maintain system stability and reliability during  traffic spikes , ensuring performance consistency across environments.   Throughput Maximization:  Deploy models optimized for peak hardware usage and maximized process throughput.  Why Join F5?   F5 empowers you to push boundaries in  AI optimization  and  high-performance engineering . Joining our team means:   Collaborating with cutting-edge technologies and hardware solutions to support real-time AI applications.   Advancing your career in a fast-paced, multidisciplinary environment focused on innovation, scalability, and problem-solving.   Driving transformative projects that deliver real-time AI reliability to global customers while maintaining cost and efficiency standards.   Working on advanced  MLOps solutions  that seamlessly scale enterprise AI systems and shape the future of intelligent deployment.  What Success Looks Like:   As an  AI Inference Engineer  at F5, success is measured by your ability to:   Combine technical expertise and problem-solving skills to deliver low-latency, scalable, and high-performing AI prediction systems.   Collaborate efficiently across cross-functional teams, participating in knowledge sharing and system refinement.   Demonstrate initiative by driving optimizations across hardware, tools, and orchestration processes, balancing immediate solutions with long-term architectural goals.   Translate complex AI and inference workflows into practical solutions that align with F5's strategic objectives.  The base pay range per annum for this position is: $176,600 - $265,000 F5 maintains broad salary ranges for its roles in order to account for variations in knowledge, skills, experience, geographic locations, and market conditions, as well as to reflect F5’s differing products, industries, and lines of busin