Optimize distributed inference across tensor, pipeline, data, context, and expert parallelism; characterize collective communication including AllReduce, AllGather, ReduceScatter, and AllToAll.
Evaluate and tune Scale-Up and Scale-Out performance, including topology-aware workload placement, network bandwidth/latency, communication-compute overlap, and scaling efficiency.
Design and execute reproducible benchmark and stress-test methodologies across model size, batch size, input/output sequence length, concurrency, accelerator count, and node count; identify saturation points and performance regressions.
Bring up and validate customer LLMs and representative open-source models; establish performance baselines, optimization plans, and competitive/production-readiness reports.
Partner with runtime, compiler, firmware, driver, platform, networking, and model teams to debug cross-layer performance issues and drive root-cause resolution.
Build Python/Bash-based profiling, benchmarking, automation, and data-analysis tooling; document reusable optimization recipes, reference configurations, and best practices.
Work directly with customers during evaluation, PoC, design-win, and production phases; convert workload requirements into clear engineering actions and communicate performance trade-offs to technical and business stakeholders.
Required Qualifications
Master’s degree or higher in Computer Science, Electrical/Computer Engineering, AI/ML, or a related technical field, with strong hands-on systems and performance engineering experience.
Strong understanding of Transformer/LLM inference, including attention, MHA/GQA/MQA, feed-forward networks, KV cache, prefill vs. decode behavior, and memory/computation characteristics.
Hands-on experience deploying and profiling LLM workloads using Python and at least one major ML framework such as PyTorch; ability to diagnose performance using traces, profiler data, logs, and system telemetry.
Experience with model optimization techniques such as quantization, batching, memory optimization, kernel/runtime tuning, and graph/compiler optimization.
Good understanding of distributed AI concepts, including multi-accelerator execution, collective communication, Scale-Up/Scale-Out fabrics, and common parallelism strategies.
Strong Linux skills and familiarity with containers and orchestration concepts (Docker/Kubernetes) used in datacenter AI deployments.
Strong scripting and automation skills in Python and Bash; disciplined approach to benchmark reproducibility and performance-data analysis.
Excellent debugging, problem-solving, and customer-facing communication skills, with the ability to work across multiple engineering organizations.
Preferred Qualifications
Experience with high-performance LLM serving stacks or optimization frameworks such as vLLM, TensorRT-LLM, SGLang, Hugging Face Transformers, DeepSpeed, or equivalent accelerator-specific runtimes.
Experience with model families such as Llama, Qwen, DeepSeek, Mixtral, or other dense/MoE LLMs and multimodal models.
Knowledge of MoE architectures, expert parallelism, token routing, and AllToAll communication performance.
Experience with RDMA/RoCE/InfiniBand or Ethernet-based AI fabrics, NUMA/topology awareness, and distributed performance analysis.
Familiarity with MLPerf Inference methodology, production observability, CI/CD performance regression testing, and capacity planning.
Experience supporting customer PoCs or production deployments of AI accelerators at server, rack, or cluster scale.
Minimum Qualifications
- Bachelor's degree in Engineering, Information Systems, Computer Science, or related field and 4+ years of Software Applications Engineering, Software Development experience, or related work experience.
- OR
- Master's degree in Engineering, Information Systems, Computer Science, or related field and 3+ years of Software Applications Engineering, Software Development experience, or related work experience.
- OR
- PhD in Engineering, Information Systems, Computer Science, or related field and 2+ years of Software Applications Engineering, Software Development experience, or related work experience.
- 2+ years of experience with Programming Language such as C, C++, Java, Python, etc.
- 1+ year of experience with debugging techniques.
Qualcomm is an equal opportunity employer. If you are an individual with a disability and need an accommodation during the application/hiring process, rest assured that Qualcomm is committed to providing an accessible process. You may e-mail disability-accomodations@qualcomm.com or call Qualcomm's toll-free number found here. Upon request, Qualcomm will provide reasonable accommodations to support individuals with disabilities to be able participate in the hiring process. Qualcomm is also committed to making our workplace accessible for individuals with disabilities. (Keep in mind that this email address is used to provide reasonable accommodations for individuals with disabilities. We will not respond here to requests for updates on applications or resume inquiries).
Qualcomm expects its employees to abide by all applicable policies and procedures, including but not limited to security and other requirements regarding protection of Company confidential information and other confidential and/or proprietary information, to the extent those requirements are permissible under applicable law.
To all Staffing and Recruiting Agencies: Our Careers Site is only for individuals seeking a job at Qualcomm. Staffing and recruiting agencies and individuals being represented by an agency are not authorized to use this site or to submit profiles, applications or resumes, and any such submissions will be considered unsolicited. Qualcomm does not accept unsolicited resumes or applications from agencies. Please do not forward resumes to our jobs alias, Qualcomm employees or any other company location. Qualcomm is not responsible for any fees related to unsolicited resumes/applications.
If you would like more information about this role, please contact Qualcomm Careers.
