Job Description

AI Platform Engineer

Company Overview

Mihira AI is a stealth-mode startup developing software and silicon for generative AI workloads. We are a group of serious GPU hardware and software nerds with decades of leadership experience from companies like NVIDIA, Intel, AMD, Apple, and others. We are funded and looking to hire our growing teams in Silicon Valley and Hyderabad, India.

Key Role Focus

We’re looking for an AI Platform Engineer who can build, test, and benchmark advanced AI hardware platforms—including edge devices, AI workstations, and AI servers—with various CPU, GPU, and accelerator configurations. The ideal candidate will design and optimize high-performance, cost-effective solutions suitable for diverse customer needs.

This role demands someone who can comfortably navigate the full hardware and software stack, tackle challenging integration issues, and deliver innovative, out-of-the-box solutions to complex hardware/software problems. A hands-on approach is essential, including assembling hardware, troubleshooting technical problems, and improving performance metrics with thoughtful, creative solutions.

Responsibilities

• Hardware Integration & Testing: Build and validate AI hardware platforms including edge devices, workstations, and servers, integrating CPUs, GPUs, and AI accelerators. Ensure stability, optimal thermal performance, and reliability.
• System Benchmarking & Optimization: Conduct comprehensive benchmarking of hardware configurations with various AI workloads. Analyze performance data and optimize system configurations, drivers, kernels, and firmware for maximum performance and value.
• Full-Stack Debugging: Diagnose hardware and software integration issues, debug Linux kernel-level problems, driver conflicts, and system performance bottlenecks, delivering stable, reliable, and high-performing platforms.
• Custom Optimization & Innovation: Develop and test creative solutions and optimizations to achieve best-in-class performance-per-dollar metrics across diverse hardware environments.
• Virtualization & Containerization: Ensure AI workloads perform optimally in containerized (Docker/Kubernetes) and virtualized environments, resolving integration and performance scaling issues.
• Documentation & Product Readiness: Maintain detailed documentation, prepare platforms for customer deployment, and ensure reliable operation in production environments.

Required Qualifications

• Full-Stack System Expertise: 5+ years’ experience in systems engineering, hardware/software integration, and performance optimization.
• Linux Kernel & Driver Skills: Extensive knowledge of Linux OS internals, kernel debugging, and custom driver development or modification.
• Hardware Assembly & Troubleshooting: Hands-on experience assembling and debugging hardware systems (CPUs, GPUs, AI accelerators, memory, storage, networking components).
• AI Benchmarking & Optimization: Proven track record of benchmarking and optimizing hardware/software configurations for AI training and inference workloads.
• System Programming Knowledge: Familiarity with low-level system tools and environments (e.g., LLVM, kernel optimization, driver stacks, scripting for automation).
• Flexibility & Hands-on Approach: Willingness and capability to physically handle hardware components, conduct on-site lab testing, and periodically travel if needed.
• Docker Container Knowledge: Expert at using docker containers and associated technologies
• Linux Kernel: Ability to customize and recompile the Linux kernel

Preferred Qualifications

• Open-Source Engagement: Contributions or active participation in open-source projects, particularly Linux kernel or driver communities.
• Accelerator Diversity Knowledge: Experience working with diverse AI accelerators and GPUs, strong familiarity with popular accelerator architectures, particularly deep knowledge of NVIDIA/AMD/INTEL GPU platforms is highly advantageous.
• Analytical & Innovative Mindset: Ability to quickly evaluate new AI hardware and software technologies, creatively leveraging them to design solutions providing significant performance, cost, or efficiency advantages.

Location

· Required: Campbell, California (Silicon Valley office).

Join Mihira AI and be part of the team shaping the future of generative AI platforms.

Job Tags

Similar Jobs