companyGimlet Labs logo

Technical Staff Member - Compiler Engineering

Gimlet LabsSan Francisco
On-site Full-time

Clicking Apply Now takes you to AutoApply where you can tailor your resume and apply.


Unlock Your Potential

Generate Job-Optimized Resume

One Click And Our AI Optimizes Your Resume to Match The Job Description.

Is Your Resume Optimized For This Role?

Find Out If You're Highlighting The Right Skills And Fix What's Missing

Experience Level

Experience

Qualifications

ResponsibilitiesDesign and implement compiler pipelines that transform high-level AI workloads into executable programs across heterogeneous hardware. Develop and refine multi-level IRs that encompass graph-level, tensor-level, and kernel-level representations. Implement partitioning and lowering strategies that effectively map workload components to suitable execution runtimes and accelerators. Support both ahead-of-time and Just-In-Time (JIT) compilation methods, including dynamic shapes and runtime specialization. Integrate new model architectures, operations, and execution patterns into the compiler framework. Collaborate closely with runtime systems, kernels, and overall systems to ensure correctness, performance, and scalability throughout the stack. QualificationsBachelor's degree in Computer Science, Computer Engineering, or related field. Solid understanding of compiler design and implementation. Experience with AI workloads and their optimization. Proficiency in programming languages such as C++, Python, or similar. Familiarity with heterogeneous computing architectures and accelerators. Strong problem-solving skills and the ability to work collaboratively in a dynamic environment.

About the job

Gimlet Labs is pioneering the first heterogeneous neocloud specifically designed for AI workloads. As the demand for AI systems increases, the industry faces critical challenges related to power, capacity, and cost with the existing homogeneous infrastructure. Gimlet's innovative solution decouples AI workloads from hardware, effectively partitioning tasks into manageable components and directing them to the optimal hardware to maximize performance and efficiency. This strategy facilitates the creation of heterogeneous systems across various vendors and generations of hardware, including cutting-edge accelerators, leading to significant advancements in performance and cost efficiency at scale.

Building on this framework, Gimlet is developing a robust neocloud for agentic workloads, enabling clients to seamlessly deploy and manage their operations via stable, production-ready APIs—eliminating the need to navigate hardware selection, placement, or intricate performance optimizations.

We collaborate with foundational labs, hyperscalers, and AI-native enterprises to support production workloads that are designed to scale to gigawatt-class AI datacenters.

We are on the lookout for a Member of Technical Staff specialized in compilers. In this position, you will engage in the core compilation infrastructure that converts high-level AI workloads into highly efficient executable programs across a diverse array of advanced hardware. Your role will include designing and implementing compiler systems that partition workloads, optimize them through various Intermediate Representations (IRs), and target multiple execution environments and accelerators.

This position is ideal for engineers who thrive on building practical systems, engaging closely with hardware, and transforming emerging AI models and execution patterns into reliable, production-ready infrastructure.

About Gimlet Labs

At Gimlet Labs, we are at the forefront of innovation, creating advanced solutions for the challenges posed by AI workloads. Our focus on developing a heterogeneous neocloud infrastructure positions us uniquely to help clients navigate the complexities of scaling AI operations efficiently and effectively.

Similar jobs

Tailoring 0 resumes

We'll move completed jobs to Ready to Apply automatically.