Senior Deep Learning Engineer

Job Title: Senior Deep Learning Engineer/AI-ML Engineer

Location: Santa Clara, California

Type: Full Time

About Us

We are team of chip design, analog and AI domain experts working towards increasing Generative AI adoption by leveraging novel in-memory computer hardware to vastly enhance the power efficiency of model deployment at the edge. We are building something entirely new to make AI ubiquitous in a responsible way widely applicable, economically viable, and environmentally sustainable.

About the Role

• The AI Framework Architect is a tech lead role building the highest level of the compiler stack, where models are ingested from PyTorch/ONNX/Keras and converted to a proprietary analog graph that can be parsed by the lower levels of the stack for hardware deployment.
• The role requires simultaneous tech lead and individual contributor (IC) work hats since the development team is highly skilled and motivated, but the complexity of the compiler requires deep expertise in graph, deep learning and software engineering principles to overcome hurdles presented by the novel chip architecture. You need to write code shoulder-to-shoulder with your team.

Responsibilities

As the tech lead for the AI framework, you will be responsible for these key facets of model deployment:

• Compiler Integration: continue to build and integrate Python-based functionality to ingest, parse and prepare model deployment to hardware
• Analytical Modeling: understand QAT, PTQ and analog-related modeling principles to optimize model representation
• Deep Learning: understand at a fundamental level how these models/networks are defined to train them for our specific workload requirements
• Graph Computation: fully comprehend at an operational level to direct graph conversion processes at lower levels of the compiler stack
• Team Development: the individual will lead a diverse, highly talented, motivated, remote team of engineers, and must be invested not only in product deliverables, but also in team members growth in terms of their professional and technical skill sets.

Candidate Requirements

The candidate will have:

• Python: 5+ years in Python development experience as an IC
• Tech Lead: have led a team of engineers in developing production-level deliverables in deep learning/AI
• Software Engineering: understand how to manage production workflows via Git/BitBucket and how to direct development efforts with Jira tasking
• Compilers: have experience working on production or proprietary compilers
• Deep Learning: have built and trained DL pipelines for models in either computer vision or NLP domain, using one or multiple DL frameworks such as PyTorch, Keras/Tensorflow, etc.
• Analytics: have an established background working on quantization techniques, stochastic processes and/or advanced numerical principles

Ideally, the candidate will have:

• Python: 8+ years in Python development
• Publication Track Record: published papers in the DL/AI in conferences or journals, and/or holding pertinent patents
• AI Accelerators: have prior work history with other AI accelerator platforms such as GPUs, TPUs, etc.

Location & Commitments

We are on a hybrid work schedule, requesting two days a week in office (Santa Clara) presence with the remaining work week being operated remotely. The hours are challenging since the development team is global, with members working from Germany, Australia and Sri Lanka. There is flexibility on the in-office presence requirement, but the tech lead role is essential for many facets of the company and needs to be readily available for meeting sessions during PST work hours.