Software Engineer, Distributed Systems
Thinking Machines LabSan Francisco, New YorkOn-Site
Needs a free Superjobs account. You'll finish on Thinking Machines Lab's Ashby page.
Skill score
Summary
This person designs and builds the core distributed systems for orchestration, scheduling, storage, and networking across thousands of GPU and TPU machines. They suit an engineer with deep systems experience who enjoys reasoning about consensus, fault tolerance, and performance under failure.
- Function
- Development
- Estimated salary
- Est. $138k–208k/yr
- Estimated for a mid-level role, ~5 yrs experience, United States. Not published by the employer.
- Work mode and location
- On-Site, San Francisco, New York
- Experience
- 5+ yrs
- Employment type
- Full-time
Job description
Overview
We're hiring a Software Engineer, Distributed Systems to design and build the core distributed systems that everything else at Thinking Machines runs on — orchestration, scheduling, storage, and networking across thousands of machines. Your work underpins both Inkling's training clusters and Tinker's serving platform, and shows up anywhere we need software to coordinate reliably at scale.
This is deep systems work. You'll be reasoning about consensus, fault tolerance, and performance under real-world failure conditions, often on problems that don't have an off-the-shelf solution.
Roles and responsibilities
- Design and build distributed systems for compute orchestration, scheduling, storage, and networking across large GPU and TPU clusters
- Develop fault-tolerant systems that keep running correctly as hardware fails, networks partition, and workloads scale
- Build the distributed storage and data orchestration layers that move and persist large volumes of training and model data
- Improve the performance and efficiency of collective communication, scheduling, and resource allocation across thousands of machines
- Partner with research and infrastructure teams to identify systems bottlenecks and design solutions from first principles
- Write production-quality code and help shape the architecture of systems used company-wide
Skills required
Top skills
- distributed systems
- Python
- Go
- C++
All skills (9)
- consensus and consistency
- fault tolerance
- network programming
- load balancing
- distributed storage
- containerization
- orchestration frameworks
- GPU and TPU clusters
- high-performance computing
Qualifications
Minimum Qualifications
- 5+ years of experience building large-scale distributed systems
- Proficiency in Python and Go, C++, or another systems-level language
- Strong understanding of distributed systems fundamentals: consensus, consistency, replication, and fault tolerance
- Experience with network programming, load balancing, or distributed storage systems
Nice to have
- Experience building distributed compute or orchestration systems for AI or ML workloads
- Fluent in containerization, orchestration, and distributed compute frameworks
- Experience with specialized hardware (GPUs, TPUs) and their integration into distributed training or serving systems
- Background at AI research labs, high-performance computing centers, or similarly demanding environments
- Published work or open-source contributions related to distributed systems or performance engineering
- Comfortable operating with high autonomy in a fast-changing, early-stage environment
Other details
- Location: This role is based in San Francisco, California or New York, New York.
Compensation
- Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $300,000 - $400,000 USD.
Benefits
- Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.
Company info
About Thinking Machines Lab
Thinking Machines Lab is an artificial intelligence research and product company that builds AI systems meant to be more widely understood, customizable, and generally capable. It offers Tinker, an API for fine-tuning open weights models, and its work is aimed at researchers and people who want to adapt AI to