Want to know if this job is worth applying to?
Boston, Massachusetts, United States
Hybrid
Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there.
Our inference stack is central to everything we ship. You'll be a core part of the team responsible for the engine layer that runs our models in production and in partner environments, and for the benchmarking infrastructure we use to evaluate our own work and verify what partners bring to us. Day to day, that means working closely with research and product, but also directly with external engineering teams.
We need someone who:
Can pick up unfamiliar tools quickly and knows how to assess whether they're worth using.
Designs AI benchmarks and holds methodology to a high standard.
Cares about inference details, understands the tradeoffs, and checks what changed across the board before calling something done.
Doesn’t consider a model port finished until you can prove the outputs are correct.
Design and build benchmark suites that cover inference performance, model quality, and knowledge evaluation across different hardware targets.
Run external partner verifications: evaluate their solutions against our benchmarks, identify gaps, and clearly deliver findings.
Port models like LFM2 onto different runtimes and frameworks, and verify correctness end-to-end.
Maintain and extend the inference engine layer built on llama.cpp, ONNX, and MLX as new model architectures emerge from research.
Make benchmark results explainable and verifiable, so internal teams and partners can trust and reproduce them independently.
Must-have:
Hands-on experience with at least one inference framework like llama.cpp, ONNX Runtime, or MLX, going beyond basic usage into internals and modification.
Experience designing and building benchmarking pipelines, including methodology, validation, and reproducibility.
Strong C++ and Python in performance-sensitive contexts.
Solid understanding of inference fundamentals: quantization, decoding strategies, memory layout, and how they interact.
Nice-to-have:
Experience porting models across runtimes and verifying numerical correctness.
Prior work with external partners or clients in a technical validation or evaluation capacity.
Familiarity with edge inference targets and the constraints that come with them.
You've ported LFM2 onto multiple runtimes and platforms, you know the model inside out, and new ports take you a fraction of the time they did at the start.
You've run multiple partner verifications end-to-end and built enough context to spot weak evaluations quickly and push back with evidence.
The benchmark suite covers inference performance and model quality across the platforms we care about, and both internal teams and partners are using it as a reference.
Compensation: Competitive base salary with equity in a unicorn-stage company
Health: We pay 100% of medical, dental, and vision premiums for employees and dependents
Financial: 401(k) matching up to 4% of base pay
Time Off: Unlimited PTO plus company-wide Refill Days throughout the year
