3 days ago
Bengaluru, IndiaMid Level / Senior
Responsibilities
- Define and evaluate AI system architectures incorporating CPUs, NPUs, GPUs, memory interconnects, and related subsystems.
- Profile and model AI workloads to optimize system performance, power efficiency, scalability, and resource utilization.
- Develop prototypes, analytical models, simulators, emulators, and proof-of-concept implementations for architectural concepts.
- Evaluate workload partitioning, scheduling, memory allocation, data transfers, synchronization, kernel selection, and accelerator utilization.
- Identify performance bottlenecks and recommend architectural improvements using workload characterization, profiling data, and benchmark results.
- Define workload-driven architectural requirements and translate them into platform specifications.
- Prepare technical proposals, architecture studies, tradeoff analyses, and performance projections for future AI platforms.
- Collaborate with hardware architects, accelerator designers, compiler teams, runtime engineers, firmware developers, performance engineers, and product teams.
Requirements
- Bachelor's or master's degree in computer engineering, electrical engineering, computer science, or a related field; minimum qualifications also allow qualifying engineering, information systems, or computer science degrees including a PhD path.
- 3-7 years of experience in a hardware-focused system optimization role, with hiring across multiple levels.
- Strong programming skills in C, C++, and Python.
- Experience in AI or machine-learning systems, computer architecture, heterogeneous computing, SoC architecture, accelerator architecture, high-performance computing, embedded systems, or system performance analysis.
- Strong understanding of computer architecture and the ability to build quantitative models and make empirically supported architecture recommendations.
- At least 2 years of academic or work experience with a programming language such as C, C++, Java, or Python.
- Preferred experience includes AI performance optimization, processor-specific optimization tools, libraries, profiling technologies, AI frameworks, transformer architectures, and optimization or deployment of large language models.
About Qualcomm
Qualcomm is a public semiconductor company headquartered in San Diego, founded in 1985, that designs and sells wireless chipsets and platforms for mobile devices, automotive, IoT, and networking, notably the Snapdragon application processors and 5G modems. It also licenses a large portfolio of cellular patents to device makers, generating revenue alongside chip sales; its technology underpins many Android smartphones and emerging automotive and edge-compute systems, and it trades on NASDAQ as QCOM.