ML Inference Engineer – Terminal-Bench Task Designer (vLLM / SGLang / TensorRT-LLM)
In short AI
Design expert-level Terminal-Bench tasks from real ML inference and model serving work, turning them into challenges AI agents must solve end to end, with a reference solution and automated verification. Paid per task, each taking about 60 to 75 minutes of expert effort, expected start early November 2026.
The ad, by section AI
- Role
- Design Terminal-Bench tasks from real ML inference work, defining the environments, writing a reference solution that runs end to end, and building automated checks that prove the result is correct, including silent failures such as numerical drift and metric bugs. Package with Docker and Git, document systems and rules, and iterate on reviewer feedback.
- Requirements
- At least 3 years of hands-on ML inference or serving work, or a PhD with active research in the subfield, with work in the last 3 years
- Trained, served or evaluated models at real scale (multi-GPU or production serving)
- Hands-on experience with PyTorch or JAX and serving stacks
- Fluency in Python and the command line, and with files, CSV/JSON, APIs or databases
- Ability to explain how ML results go wrong without errors and how to catch them
- Availability to start by early November 2026
- Preferred
- AI lab or LLM infrastructure experience, contributions to PyTorch, vLLM or SGLang, GPU profiling, Docker and Git, automated testing
- Conditions
- Contract: task-based service provision (paid per task)
- Salary: not stated
- Hours: about 60 to 75 minutes per task; weekly hours not stated
- Location: remote
- Benefits
- none stated
Overview We are building a benchmark that tests how well AI agents handle real expert work in a terminal. We are hiring ML inference and serving experts to turn real, paid tasks from their jobs into expert-level Terminal-Bench challenges. Each task includes the environment, a working solution, and an automated check that proves the result is correct. ***THIS IS A TASK BASED COMPENSATION MODEL.…
The source asked us not to reproduce the full text here: the complete ad is at the original link.
How to apply
Application: mention the word ASTONISHED and the tag ROTUuOTMuMzEuMTgy. Expected start early November 2026. Process: vetting task instead of an interview, then per-task reviewer feedback. AI
Prepare your application
A cover letter and a short pitch for this job, written from your profile, and what to learn next.
Sign in and create your profile from your CV. It is private: we share it with no one.
Something wrong or unlawful in this ad? Report it
Similar jobs
- Operations Expert with Python for AI Supply Chain Task DesignToptal
- Machine Learning Engineer — Inference OptimizationFeatherless AI
- CAD scripting ExpertToptal
- ML Engineermicro1
- ML EngineerEVB
- 🌍 ML Engineer, Remote- ContractXperteez Technology
Salary compared ⓘ
86 jobs · Systems analysts · remote
see the best paid →Training for this job
- AI267 - Developing and Deploying AI/ML Applications on Red Hat OpenShift AIRumosblended · 21 h · 953 € + IVA · starts 01/01/2027Asked for in this ad: AI / machine learning
- AI267: Developing and Deploying AI/ML Applications on Red Hat OpenShift AI (B-Learning)Galileuself-paced · flexible hours · 40 h · 1905 €Asked for in this ad: AI / machine learning
- Inteligência Artificial & Data ScienceCESAE Digitalonline · evenings · 75 h · freeAsked for in this ad: AI / machine learning
Languages AI
Published on (2)
- RemoteOK seen 6 Nov – 21 Nov
- We Work Remotely seen 16 Nov – 20 Nov
First seen on 6 Nov, 21:00 · last seen on 20 Nov, 20:35