Compiler Director

Title: Compiler

Location: Greater Bengaluru Area (Hybrid)

About Company:

We are a leader in advanced AI hardware and software systems for edge-to-cloud computing. Our robust and scalable next-generation in-memory computing technology provides orders-of-magnitude higher compute efficiency and density compared to todays best-in-class solutions. The high-performance architecture is coupled with seamless software integration and will enable the immense potential of AI to be accessible in power, energy, and space constrained applications. We were launched in 2022 and is led by veteran technologists with backgrounds in semiconductor design and AI systems.

Key Responsibilities

Technical Leadership : Define the compiler roadmap and architecture, driving innovation in graph compilation, lowering flows, and performance optimization.

New Model Enablement : Own end-to-end bring-up for new model families (e.g., Llama, MoE) from FX capture through runtime execution.

Frontend & IR Development :

o Oversee development of modules such as llama.py, ensuring alignment with compiler FX expectations.

o Lead FX graph transformations and lowering to EnCharges EAI primitive dialect (pattern library).

Kernel & Ops Development : Direct kernel library integration, new kernel creation, and custom NPU RV ops to extend hardware capabilities.

Feature Ownership : Deliver compiler support for advanced features like device-to-device (D2D) communication and attention optimizations.

Cross-Team Collaboration : Partner with hardware, quantization, runtime teams to ensure feature completeness and performance parity across new models.

Quality & Execution :

o Establish robust QA, validation, and regression testing for compiler releases.

o Ensure timely, high-quality software delivery with focus on reliability and maintainability.

People Management :

o Build, mentor, and scale a high-performing team of compiler and software engineers.

Qualifications:

15+ years in compiler or systems software development, with 5+ years leading engineering teams.

Expertise in AI/ML graph compilers (e.g., Torch-MLIR, TVM, XLA, Torch-FX).

Strong understanding of neural network architectures, quantization, and hardware acceleration.

Proven success driving complex, cross-functional software initiatives from design to release.

Experience with MLIR/LLVM, hardware-software co-design, and open-source compiler contributions.

Exceptional leadership, communication, and collaboration skills.

Contact

Sumit S. B.

Mulya Technologies

"Mining the Knowledge Community"