Machine Learning Engineer, Gensyn
Research
+ equity
More information about location
Hyperscale machine learning compute protocol
Job no longer available
Hyperscale machine learning compute protocol
1-20 employees
Job no longer available
+ equity
More information about location
1-20 employees
Company mission
To enable machine learning engineers to train AI models at a higher scale and lower cost.
Role
Who you are
- Huge-scale model training - previously trained enormous models (e.g. LLMs) over distributed infrastructure
- Infrastructure experience - previously provisioned and/or configured large-scale infrastructure for ML training tasks and used communication backend tools like NCCL, GLOO
- Background in relevant theory - i.e. computer science, machine learning, algorithms, distributed systems
- Strong willingness to learn Rust - as a Rust by default company, we require that everyone learns Rust so that they have context/can work across the entire codebase
- Highly self-motivated with excellent verbal and written communication skills
- Comfortable working in an applied research environment - with extremely high autonomy
Desirable
- Rust experience - systems level programming experience in Rust
- Compiler-level experience - previously worked at the lower levels of the ML stack - on compilers or hardware implementation
- Quantisation / constrained optimisation - previously worked on overcoming bottlenecks in ML training from hardware or communication constraints
- Decentralised model training frameworks - exposure to frameworks like Moshpit SGD that effectively optimise models despite high latency, heterogeneous, compute environments
- Open source work - experience working with large open source codebases (ideally as maintainer)
What the job involves
- Implement novel research - build out newly proposed model distribution methods and low-level primitives to train huge models over a completely new infrastructure
- Test SoTA frameworks - implement and test state-of-the-art distributed training frameworks over various settings including heterogeneous and low bandwidth GPU networks
- Stay up-to-date on distributed learning developments - follow the research and trends in both industry and academia regarding distributed learning
- Research support - work with the research team on wider issues concerning distributed ML (e.g. asynchronous learning)
- Follow best practices - build in the open with a keen focus on designing, testing, and documenting your code
- Write & engage - contribute to technical reports/papers describing the system and discuss with the community
Our take
The concept of "deep learning" evolved decades ago and is required for AI-based applications, with the necessitates of its development based on patterns found in the human brain. Most AI developers, though, cannot afford the computing power needed to run these AI applications. Instead, they're forced to rent from cloud computing platforms.
Gensyn is a London-based blockchain start-up that is building cost-efficient and decentralised deep learning and machine learning compute networks, allowing deep and machine learning models to be trained at a new scale and at a cheaper cost. At a time when the popularity of artificial intelligence is soaring, the potential for such an offering is high, definitely making the company one to watch.
Founded in 2020, Gensyn's long list of investors include Eden Block - a Web 3.0 infrastructure VC - as well as founders from major blockchain organisations. Its latest raised, headed by Andreessen Horowitz, is cited to support the expansion of Gensyn's team, as well as help with the finalization and launch of its product.
Steph
Company Specialist at Welcome to the Jungle
Company
Funding (last 2 of 3 rounds)
Jun 2023
$43m
SERIES A
Mar 2022
$6.8m
SEED
Total funding: $50.9m
Company benefits
- Fully remote - we hire between the West Coast (PT) and Central Europe (CET) time zones
- Flat hierarchy and an autonomous environment
- Share of equity and token pool
- 4x all expenses paid company retreats around the world, per year
- Whatever equipment you need
- Paid sick leave
- Private health, vision, and dental insurance - including spouse/dependents 🇺🇸
Company HQ
Westminster, London, UK
Leadership
Ben Fielding
(Co-founder)Previously CEO and Co-founder of Fair Custodian and a Director at Research Analytics, as well as a Machine Learning Consultant at Ricochet AI.
Harry Grieve
(Co-founder)An angel investor for companies like Thymia, Altitude Labs, Catapult, and Modulus Labs. Worked as a Data Research Director at Cytora prior to Gensyn.
Share this job