Are you excited by creating groundbreaking products from the ground up which
hold the potential to transform industries and impact people's lives? Are you
interested in enabling a future where AI is safe, efficient and ubiqitous? Are
you looking for a career change and to be a part of an exceptional team? If
so, we cannot wait to hear from you! (seriously, im not kidding, reach out
asap!)
We are in the midst of experiencing a massive shift in how we write software,
from Software 1.0 (traditional) to Software 2.0 (data centric). This new
paradigm will allow us to realize the true benefits of AI by leveraging deep
learning and reinforcement learning. However, to fully realize these benefits
we need a ground up reimagining of computer architectures to make them data
and AI centric in order to bring down the eye-watering energy requirements to
power these models, while making the lives of application engineers easier.
At Lemurian Labs our purpose is to create better hardware and systems using a
software-first approach that enables those trying to build the future of AI to
do more than they ever thought possible, to free them from their computational
burdens, and make it easier to train and deploy models into production so that
society can benefit from this immensely powerful technology.
Our SPU (Spatial Processing Unit) is the most powerful and efficient
programmable digital processor ever created, and it dramaticallt reduces the
capex and opex of AI.
Our work cuts across every layer of the engineering stack - from the hardware,
runtimes, compilers, kernel optimization, algorithm development, to software
architecture. If you find performance engineering fascinating and the
possibility of making a lasting impact on a field appealing, we want to hear
from you.
We need our engineers to be curious, versatile, creative and display
leadership qualities and be enthusiastic to take on new problems and across
the full stack as we continue to push boundaries.
As a Sr Compiler Engineer you will be responsible for developing compiler
optimizations for our state of the art optimizing compiler which generates
highly optimized code for our target hardware. We are a small team that
prioritizes cross collaboration and values thinking differently. Therefore, a
commitment to collaborative problem solving, sophisticated design, and the
creation of quality products is essential. Join us in our mission to make a
tremendous and lasting impact and help bring forth the future of AI.
Responsibilities:
* Design, develop, maintain and improve our next-gen SPU compiler
* Propose improvements to and expand our IR with respect to new advancements in machine learning model architectures
* Design and implement new passes and innovative scheduling techniques
* Use the latest techniques in parallelization and partitioning to automate generation and exploit highly optimized kernels
* Rapid prototyping and data driven exploration of new ideas
* Benchmark and analyze the outputs produced by our optimizing compiler on our SPU hardware
* Work closely with both the hardware and software teams to understand the evolving needs of ML engineers and drive improvements in architecture
* Build tools to collect and analyze performance bottlenecks
Qualifications:
* A degree in computer science, computer engineering, electrical engineering, or equivalent practical experience [Applicants with Masters or PhD will be given preference]
* 2+ years of experience working with compilers (especially compiler backends) and retrargeting
* 5+ years of experience with C/C++ (C++14 or newer) and Python
* Knowledge of functional programming
* Experience with loop optimizations (vectorization, unrolling, fusion, parallelization, etc)
* Experience with FPGAs or CGRAs
* Knowledge with DL frameworks such as Tensorflow or PyTorch is desirable, but not essential
* Working knowledge of LLVM, MLIR and polyhedral models
* Exposure to ONNX will be an asset
* [If you're a strong programmer, have a curious mindset and are willing to learn, we will teach you everything you need to know on ML/DL/RL and compilers to be effective in the job]
Personal Attributes:
* Detail oriented
* Strong written and oral communication, and able to write clear and concise documentation
* Motivated to see projects to completion
* Team first attitude
* Self starter
* Ability to pick up new skills and adapt in fast changing environments
Lemurian Labs welcomes applications from all sections of society. We are
committed to equal employment opportunity regardless of sex, race, religion or
belief, ethnic or national origin, disability, age, citizenship, marital,
domestic or civil partnership status, sexual orientation, gender identity,
pregnancy, maternity or related condition (including breastfeeding) or any
other basis as protected by applicable law. If you have a disability or
additional need that requires accommodation, please do not hesitate to let us
know.
Lemurian Labs focuses on Semiconductors, Robotics, Artificial Intelligence,
Internet of Things, and Drones. Their company has offices in Toronto. They
have a small team that's between 11-50 employees.
You can view their website at <http://www.lemurianlabs.com> or find them on
[LinkedIn](https://ca.linkedin.com/company/lemurianlabs).
