Qualcomm · Automotive Systems

Arjit
Chauhan.

Senior Systems Engineer

Specializing in Linux kernel internals, QNX Neutrino IPC/RTOS, and hypervisor scheduling for next-generation automotive platforms.

scroll

I'm a Senior Systems Engineer at Qualcomm, specialising in automotive SoC performance, Linux kernel internals, QNX Neutrino IPC/RTOS, and hypervisor scheduling.

My expertise includes configuring JTAG-level hardware counter tracing, solving queueing-model problems for scheduler contention, and automating concurrent characterisation pipelines, enabling deterministic performance in automotive domains.

I enjoy engineering profiling toolkits directly interfacing with hardware counters as well as taking on complex full-system integration challenges across CPU, GPU, DDR, and NPUs.

Linux kernel internals QNX Neutrino IPC Hypervisor (KVM/QNX) JTAG debugging Automotive SoC LLCC/DDR hardware counters C++17 Python
Role Senior Systems Engineer
Company Qualcomm Inc.
Location Bengaluru, KA, India
Education M.Tech CSE — NIT Jalandhar (2021–2023)
B.Tech (Hons) CSE — RTU (2017–2021)

Areas of Focus

Building and optimising systems where hardware meets software at the edge of real-world performance.

Systems Engineering

Deep expertise in Linux kernel internals, QNX Neutrino IPC/RTOS, hypervisor scheduling (KVM/QNX Hypervisor), and SoC bring-up.

Observability

Advanced granular profiling involving perf_events, flame graphs, and extracting LLCC/DDR hardware counters.

Hardware & Verification

Bringing up automotive SoC (Snapdragon SA-series platforms), DDR5/LPDDR5 controller analysis, and conducting Vmin characterization pipelines.

Development Tooling

Developing low-level profiling toolkits in C++17 and automating complex generation pipelines via production-grade Python and Bash.

Experience & Education

Qualcomm Inc.
Jan 2026 — Present
Senior Systems Engineer
  • Profiled and root-caused CPU runqueue saturation across concurrent IVI/ADAS/Flex EUC workloads on automotive QNX targets; applied queueing-model analysis to identify scheduler contention as primary latency driver, enabling 80%+ sustained system utilisation with deterministic deadline compliance.
  • Diagnosed hypervisor scheduling jitter on QNX/Android Automotive dual-OS targets; isolated VM exit latency as root cause via JTAG-assisted tracing; tuned vCPU affinity policy reducing worst-case inter-VM preemption latency.
  • Identified LLCC thrashing pattern during peak CPU+GPU+NPU concurrent EUC load via hardware performance counter analysis; proposed and validated cache-partitioning policy that recovered ~10% memory bandwidth utilisation across DDR controllers.
Qualcomm Inc.
July 2023 — Dec 2025
Systems Engineer
  • Owned performance characterisation for 5+ automotive SoC SKUs from bring-up through customer qualification; used JTAG-level hardware counter tracing to isolate silicon-level bottlenecks that blocked 3 customer tape-out sign-offs.
  • Architected a 2KLoC Python framework automating concurrent Vmin characterisation across CPU, GPU, DDR, and NSP subsystems; reduced full SoC voltage-characterisation cycle from 7 months to under 2 months across 3 successive SoC generations.
  • Collaborated with design, validation, and marketing teams to support system-level optimization and go-to-market efforts.
Qualcomm Inc.
Aug 2022 — June 2023
Interim Engineering Intern
  • Engineered a BERT-based TAPAS model chatbot supported seamless aggregation and sequential conversational queries, enhancing user interactions.
  • Developed web dashboards using React and Flask, delivering intuitive data presentation and enabling efficient component comparisons.
  • Designed and implemented an in-house Priority Request Management system to streamline the management and prioritization of inter-department requests.
NIT Jalandhar
Aug 2021 — 2023
M.Tech. Computer Science and Engineering
  • Major Coursework: Advance Data Structures and Algorithms; Machine Learning; Decision Support Systems; Cloud Computing; Advanced Databases and Data Mining.
RTU Kota
Aug 2017 — 2021
B.Tech.(Hons.) Computer Science and Engineering
  • Major Coursework: Theory of Computation; Database Management System; Operating Systems; Data Structures and Algorithms; Computer Networks.

Selected Works

A selection of projects spanning systems automation, performance engineering, and applied ML.

QNX/Linux SoC
Performance Toolkit

Engineered a custom profiling toolkit to automate flame graph generation and interface directly with hardware counters via perf_event_open() for granular IPC latency benchmarking across QNX and Linux environments.

C++17 Python perf_events
LLCC Allocation
Optimizer

Automated the end-to-end LLCC partition sizing problem: profiled CPU, GPU, and Multimedia client cache working sets via hardware counter sweeps, fit marginal-gain curves per client, and solved a constrained allocation over the total LLCC budget to minimise DDR bandwidth while satisfying per-client latency SLAs.

Programmatically applied the computed SCID configuration to the target board and ran the full EUC benchmark suite, producing before/after reports on DDR bandwidth utilisation and cache hit rates across subsystems.

Python hardware performance counters QNX/Linux
Health Insurance Recommendation
using Probabilistic Data Structures

Chauhan, A., Singh, A. "Health Insurance Recommendation using Probabilistic Data Structures" 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India, 2023, pp. 1-6 DOI: 10.1109/ICCCNT56998.2023.10307603.

Publication ICCCNT
Community Volunteer
Google Developer Groups

Data Science Facilitator, GDSC JGI: Volunteered multiple workshops on basic as well as emerging Technologies (Sep 2019 - Aug 2021).

Core Team Member, GDG Jodhpur: Volunteered in DevFest - 2019 (Aug 2018 - Aug 2020).

Volunteering GDSC GDG
[LATENCY: 0.04ms]
[SYNC: ACTIVE]
[ARCH: ARM_Oryon]
[EXP_ID: 0x9F28]

Augmented Workflows

Technical experiments leveraging agentic AI to push the boundaries of research and visualisation.

EXPERIMENT_VIEWER v1.0.4
STABLE
Thought

An experimental exploration into cognitive patterns and historical systems. Visualized via a bespoke 3D/2D interface designed to reveal latent connections within high-dimensional datasets.

Launch Environment →
PROCESSOR Neural_Engine
VISUAL_ENGINE ThreeJS_GL
DATA_SOURCE Historical_CSV
STATUS OPTIMIZED

Let's Talk

Open to conversations about systems engineering, automotive AI, or anything interesting.

arjitc12@gmail.com