Software Engineer, Production Inference (Distributed Inference)
Thinking Machines Lab · San Francisco ·
- Work mode
- Onsite
- Employment
- Full time
- Category
- Software engineering
- Experience
- 3+ years
- Visa sponsorship
- Yes
- Salary
- USD 350,000 – 500,000 / year
Thinking Machines Lab · San Francisco ·
Build and operate the distributed production inference systems that serve Thinking Machines' frontier models (Inkling, Inkling-Small) and Tinker — request routing, batching, multi-node GPU serving, and latency/throughput optimization (KV cache, speculative decoding, quantization). Systems-heavy role needing strong Python/C++/Rust and distributed systems experience, with on-call participation.
The mission of Thinking Machines is to build AI that extends human will and judgment. We are training frontier models with Inkling, developing Tinker to let people make models their own, and crafting interfaces that broaden human-AI communication. We believe the future worth building is human, and we're hiring people who want to build it.
We're hiring a Software Engineer to build and scale the distributed production inference systems that serve Inkling, Inkling-Small, and Tinker in production. You'll own the systems that turn trained models into fast, reliable, cost-efficient services — from request routing and batching to multi-node serving and GPU utilization at scale.
This is a systems-heavy, production-first role. You'll work closely with research and infrastructure teams to translate rapidly evolving model architectures into serving systems that meet real-world latency, throughput, and reliability requirements, and you'll be on the front line when production inference systems need to scale, recover, or improve.
Design, build, and operate distributed infrastructure for large-scale model serving, including request routing, load balancing, batching, and multi-node coordination
Optimize inference latency and throughput in production, including work on KV cache management, continuous batching, speculative decoding, and quantization
Build and maintain high-concurrency serving systems with strong uptime, low tail latency, and deep observability
Benchmark, tune, and extend inference engines to support new model architectures as they move from research into production
Partner with research and infrastructure teams to translate emerging model designs into production-ready serving systems
Build tooling for tracing, debugging, and resolving issues across the serving stack, from orchestration down to GPU kernels
Participate in on-call rotation to support production inference systems
3+ years of experience building and operating distributed systems in production
Strong systems programming skills in Python, C++, Rust, or similar languages
Experience with production infrastructure at scale: reliability, observability, and performance under real-world load
Solid understanding of networking, concurrency, and distributed systems fundamentals
Experience with LLM inference engines such as vLLM, SGLang, or TensorRT-LLM
Familiarity with GPU programming (CUDA) or low-level performance optimization
Experience with model parallelism, tensor/pipeline parallelism, or other distributed inference techniques
Track record of operating large-scale production systems with strict latency and uptime requirements
Experience with Kubernetes or similar orchestration systems for GPU workloads
Contributions to open-source ML systems or inference infrastructure projects
Location: This role is based in San Francisco, CA.
Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $500,000 USD (placeholder — verify against current internal bands before publishing).
Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.