Rahul Kukreja

DevOps Engineer

Bengaluru, Karnataka, India13 yrs 1 mo experience
Highly Stable

Key Highlights

  • Expert in performance modeling for GPUs and SoCs.
  • Proficient in multi-core programming with CUDA and MPI.
  • Strong background in embedded systems and ECU simulation.
Stackforce AI infers this person is a Semiconductor and Automotive expert with a focus on performance modeling and embedded systems.

Contact

Skills

Core Skills

Performance ModelingGpu Performance OptimizationPerformance AnalysisModelingEmbedded Systems

Other Skills

C#C++CUDAComputer ArchitectureCycle Accurate Performance Model SimulatorDesign Space ExplorationEmbedded CLaTeXLinuxMPIMatlabMicrocontrollersMicrosoft OfficeObject-Oriented Programming (OOP)OpenMP

About

* Intrigued by the world of Computer Architecture. * Experienced, goal-driven and self-starting engineer requiring minimal supervision while performing tasks. Capable of working equally efficiently either independently or as part of a team. * Constantly demonstrated the effective application of technical, presentation and communication skills. * Strong C and Object Oriented Programming (C++, C# and Java) skills. * Multi-core programming using CUDA, MPI, P-threads, OpenMP and SSE. Focus areas: * Performance modelling * Computer Architecture

Experience

13 yrs 1 mo
Total Experience
3 yrs 8 mos
Average Tenure
2 yrs
Current Experience

Amd

SMTS

Apr 2024Present · 2 yrs · Bengaluru, Karnataka, India · Hybrid

Qualcomm

3 roles

Staff Engineer

Promoted

Dec 2021Apr 2024 · 2 yrs 4 mos

  • Working in the Graphics systems team. Involved in the following activities:
  • Development of a cycle accurate performance model simulator (SystemC based) for the analysis of future generation GPUs.
  • Correlation of performance data obtained from the above model with Pre-Si data.
  • Analysis of graphics workloads to identify architecture limitations and provide appropriate solutions.
  • Working closely with architects to design and drive new feature development for optimal GPU performance and power consumption.
  • Development of tools and scripts (Perl and Python based) that help in the analysis and comparison of performance counters obtained from different platforms.
  • Parallelly, drove the development of a performance simulator which allows co-simulating blocks coded in a high-level language (SystemC/C++) with blocks more accurately coded in RTL. This model allows in making feature changes and observing its impact on overall GPU performance with small turnaround times
Performance ModelingCycle Accurate Performance Model SimulatorSystemCPerformance AnalysisGPU Performance OptimizationPerl+1

Senior Lead Engineer

Dec 2019Dec 2021 · 2 yrs

Senior Engineer

Aug 2017Dec 2019 · 2 yrs 4 mos

Intel corporation

2 roles

SoC Design Engineer

Feb 2016Aug 2017 · 1 yr 6 mos · Malaysia

  • Worked as a front-end RTL design engineer and integrator for different System on Chip (SoC) projects.
  • Designed, tested and integrated an interrupt controller for the imaging unit of a SoC.
  • Owned the integration of fabrics (communication infrastructure) in a client based SoC.
  • Involved in the unified power format (UPF) design of the above components.

Intern

Sep 2014Aug 2015 · 11 mos · Netherlands

  • Involved in the modelling and analysis of a ring interconnect (RI) fabric that will be used in future generation Image processing units.
  • 1) Developed C++ models of various components of the RI in accordance with proprietary communication protocols.
  • 2) Implemented a flexible C++ model used to test and monitor the performance of the RI using real world scenarios.
  • 3) Used the models for cycle accurate simulations to perform Design Space Exploration and determine optimal configurations of the RI for different traffic scenarios.
  • 4) Created Perl scripts to automate pre and post simulation processing.
C++PerlDesign Space Exploration

Robert bosch engineering and business solutions ltd.

Software Engineer

Jul 2011Jul 2013 · 2 yrs · Bangalore, India

  • My job description entails developing tools that aid in simulating the Engine Control Unit (ECU) of a car, to test various functionalities such as anti-lock braking (ABS), Hill Hold Control (HHC), DSC and more.
  • 1) I was given the opportunity to design and develop a measurement technique that is used to measure the signal data entering or leaving the ECU during Software in loop (SiL) simulations (Open loop and Closed loop).
  • 2) I also developed tools that are used to speed up test cases. For example, I developed a tool that is used to compare the signals of an ECU after different runs of simulations; thereby eliminating unnecessary time and effort.
  • 3) The Make Tool Chain (MTC) is a set of tools that are used to build the ECU software. Depending on the type of build required these tools are called in a particular order. Developed a platform (Using Windows PowerShell) that:
  • Builds one or two different projects with specified MTC versions.
  • Simulates the projects and then compares the simulation output files.
SimulationTool DevelopmentSignal ProcessingEmbedded Systems

Education

Delft University of Technology

Master of Science (MSc) — Embedded Systems

Jan 2013Jan 2015

PESIT,bangalore

Bachelor's degree

Jan 2007Jan 2011

JAIN College

Pre University

Jan 2005Jan 2007

Sindhi High School

High School

Jan 1995Jan 2005

Stackforce found 100+ more professionals with Performance Modeling & Gpu Performance Optimization

Explore similar profiles based on matching skills and experience