FY26 Intern - ML Compiler Research Intern (Megakernels) - 6 months, Strasbourg, France

Location: Strasbourg, France

Department: Interim Intern

Work Location: onsite

Company:

Job Area:

General Summary:

Opportunity Overview:

Join Qualcomm’s Compiler Labs in Strasboug France for a 12–14 week summer internship (May–August 2026, with some flexibility) focused on the approach of programming ML accelerators using a “megakernel” approach. Current ML compiler tools for frameworks such as ONNX, Pytorch, start with representations of the ML model as graphs and then work to lower the graph into calls to catalogs pre-written operator kernels, which are then scheduled under resource constraints of processor, memory and communication bandwidth. In some cases, the compiler can execute fusion in various forms to optimize communication. In the megakernel approach, a kernel is a large section of the graph represented and optimized jointly to allow for more coherent optimization at well-aligned representations of that section of the model. Interns will build within existing processes for making and evaluating megakernels—targeting the Hexagon NPU, and with opportunities to address Arm SME. This is a unique chance to work on the leading edge of ML compiler technology, contributing to GenAI model optimization across a breadth of hardware markets—edge and cloud.

Representative Projects:

• Extend and generalize existing megakernels to represent new ML model algorithm capabilities.
• Extend and create new automatic approaches for generating megakernels
• Perform performance measurement and profiling to assess approaches and guide towards development of new ones.
• Focus on measurable improvements in full model compilation and deployment, leveraging existing optimization infrastructure.

Toolchain & Environment:

MLIR (Affine/Linalg/Transform), LLVM, PyTorch 2.x (FX/Inductor), OpenAI Triton DSL, C++17, Python.

Qualcomm Hexagon toolchain, Adreno GPU dev tools, performance profilers.

Hands-on access to modern AI accelerator toolchains and developer hardware.

Mentorship & Experience:

Weekly 1:1 mentorship with senior compiler engineers and polyhedral experts. Mid-summer review and access to internal AI/compiler learning sessions.

Culminating demo/talk at the end of the internship.

Publication & Open Source:

Positive publication stance (e.g., MLIR Workshop); intent to open-source results where feasible, subject to Qualcomm review and approval.

Support for academic internship reports after legal review.

Eligibility & Application:

PhD or research-oriented MS students in CS/CE/EE, with advanced compiler coursework and research experience.

Graduation target: 2027 preferred.

CPT/OPT accepted; finalists may undergo export-control screening per policy. Competitive compensation, with housing and relocation support provided.

To apply: Apply your resume (and optionally one paper or code sample) plus a brief statement of research fit.

Why Qualcomm?

Work at the intersection of intelligent computing and low-power performance, shaping how ML and GenAI run on real devices. Qualcomm’s breadth of hardware markets—spanning edge to cloud—makes ML and GenAI performance central, with deep overlap between information theory and machine learning.

Note: Interns follow Qualcomm confidentiality and open-source contribution policies; publications and upstreaming require prior approval.

Summer 2026 ML Compiler Research Intern (Megakernels)

Location: Strasbourg, France | On-site, 5 days/week

*References to a particular number of years experience are for indicative purposes only. Applications from candidates with equivalent experience will be considered, provided that the candidate can demonstrate an ability to fulfill the principal duties of the role and possesses the required competencies.

Qualcomm expects its employees to abide by all applicable policies and procedures, including but not limited to security and other requirements regarding protection of Company confidential information and other confidential and/or proprietary information, to the extent those requirements are permissible under applicable law.

To all Staffing and Recruiting Agencies: Our Careers Site is only for individuals seeking a job at Qualcomm. Staffing and recruiting agencies and individuals being represented by an agency are not authorized to use this site or to submit profiles, applications or resumes, and any such submissions will be considered unsolicited. Qualcomm does not accept unsolicited resumes or applications from agencies. Please do not forward resumes to our jobs alias, Qualcomm employees or any other company location. Qualcomm is not responsible for any fees related to unsolicited resumes/applications.

If you would like more information about this role, please contact Qualcomm Careers.