Machine Learning Engineer (ML for EDA)

Position Overview We are seeking a Machine Learning Engineer to improve the models that power our agentic AI platform. You will work on post-training, fine-tuning, and evaluation of large language models for chip design and verification tasks, taking research ideas from experiment to production. Your work will directly impact the quality and reliability of AI agents used by leading semiconductor companies.

Key Responsibilities

• Train, fine-tune, and post-train LLMs for chip design and verification tasks, using techniques such as supervised fine-tuning and reinforcement learning.
• Build and maintain data pipelines for training, including data collection, cleaning, and synthetic data generation.
• Design evaluation frameworks and benchmarks that measure model accuracy, reliability, and agent performance on real engineering workflows.
• Run experiments, analyze results, and iterate quickly to improve model quality.
• Collaborate with research scientists, systems engineers, and EDA experts to bring new models and methods into production.
• Keep up with emerging research and open-source work, and apply promising techniques to our models.

Qualifications

• B.S., M.S., or PhD in Computer Science, Electrical Engineering, or a related field (or equivalent experience).
• Hands-on experience training or fine-tuning large language models, including SFT, RLHF, or other post-training methods.
• Strong proficiency in Python and PyTorch; experience with Hugging Face Transformers or similar libraries.
• Experience building evaluation harnesses and data pipelines for ML systems.
• Solid understanding of modern LLM architectures and training dynamics.
• Familiarity with distributed training frameworks and GPU workflows is a plus.
• Experience with agentic workflows, code generation models, or hardware design (RTL, Verilog, SystemVerilog) is a plus.
• Self-directed problem solver who is interested in working on ambitious, open-ended challenges.

Nice to have

• Proficiency in C++ and CUDA.
• Understanding of GPU architecture, memory hierarchies, and parallel computing.
• Hands-on experience with inference frameworks such as SGLang or vLLM.
• Experience deploying and optimizing LLMs in production, such as model serving, batching strategies, distributed inference, or quantization.
• Systems-level debugging and profiling skills, from CUDA kernels to application logic.
• Familiarity with distributed computing frameworks (Ray, multi-node training and inference).
• Experience with agentic workflows, code generation models, or hardware design (RTL, Verilog, SystemVerilog).

Why Join Us

• Work on cutting-edge LLM post-training and evaluation problems with real-world production impact.
• Working with research scientists, software engineers and EDA veterans (including a former Synopsys VP) on a flat team that rewards good ideas, and helping close the loop between front-end and back-end chip design.
• Access to substantial GPU compute resources for experimentation.
• Shape the quality of AI systems used by leading semiconductor companies.