Our vision is to transform how the world uses information to enrich life for .
Micron Technology is a world leader in innovating memory and storage solutions that accelerate the transformation of information into intelligence, inspiring the world to learn, communicate and advance faster than ever.
JR62815 Systems Performance Engineer
Micron's Compute and Network Business Unit (CNBU) is looking for a strong principal engineer or member of technical staff (MTS) to drive the client and data center workload engineering activities in Hyderabad. There is the urgent need to understand how data center systems and client systems (AI-PC) are composed using Micron's memory hierarchy so we can better position our products based on workloads, derive richer requirements for future product design, produce marketing collaterals based on workload performance data and create thought leadership and PoCs. To this end, the candidate will contribute to several areas in workloads performance characterization, systems architecture analysis, future design via systems performance analysis/simulation/emulation and evaluate & prototype software systems where appropriate.
Responsibilities include, but not limited to:
- Analysis and characterization of server or client laptop workloads in several areas such as AI/ML inference, gaming and rendering.
- Characterization of data center server systems, AI-PC systems and emerging accelerator technologies in concert with the memory hierarchy therein.
- Understanding workloads using several application, overall system and memory subsystem based KPIs (including throughput, performance, power, memory bandwidth, latency, battery life, microarchitectural usage, etc.)
- Defining Memory Systems architectural requirements with market competitive analysis for Client (laptop) or server platforms
- Innovating new solutions to sophisticated System level problems by collaborating with Engineering and Product Management teams with required System Level expertise i.e., SW and HW
- Developing, evaluating and optimizing new memory architectures that continuously improve cost, performance, power efficiency of Client DRAM products
- Defining cost, performance, power metrics for Client-Edge platforms and quantifying solutions with justifying analysis
- Evaluating and demonstrating architecture trade-offs that impact memory subsystem performance
- Own performance analysis end to end, by coming up with a value proposition to showcase a product, collecting requirements from stakeholders, putting together a system architecture, compiling a set of key workloads in AI/ML and gaming, conducting experiments and analyzing the data, deriving key insights and communicating it to stakeholders and customers.
Successful candidate for this position will have a subset of the following:
- Must have a BS or Master's degree or PhD in Computer Science or Computer Engineering
- 10+ years of experience in Systems architecture, DRAM or NAND IPs or SoC or IP Design & Performance Analysis for Client/Server/Mobile systems
- Experience in CPU, GPU, AI Processors, DDR, or systems development
- Solid understandingofComputer systems (client or server) and Memory or Storage Architecture
- Experience in critical metric (Power, Performance, Thermal) analysis and in crafting/improving/using various profiling (Latency, Power, etc.,) tools
- Experience with performance analysis, modeling, emulation or simulation
- Software development experience in building simulators for processor or memory subsystems
- Experience in leading highly skilled teams both technically and organizationally
- Excellent communications skillswith theability to articulate results
- Strong leadership skills and multi-functional collaboration
- Self-motivated and significant problem-solving skills
Preferred Skills:
- 15+ years of systems architecture experience in client/server HW/SW
- Application workload performance analysis using several client/server systems performance tools
- Strong working knowledge of DRAM and other memory devices
- Working Knowledge of DDR or LPDDR3/4/5 PCIe
- Exposure to Custom HW platform development
- Strong background in accelerating AI models in a server/PC environment
- Strong background in using/developing simulators/emulation/analytical modeling for performance analysis
