Remote AI training and data labeling jobs

Materials scientists write connected, multi-step scientific coding problems and tested Python solutions for Turing's SciCode set. 8-week contractor assignment, 40 hours a week, 12 eligible countries. No rate published.

Biology graduates author linked scientific coding problems with tested Python solutions for the SciCode training set, which the ad says is run in partnership with NVIDIA. 8-week contract, 40 hours a week, 12 countries. No rate published.

Chemistry graduates write multi-step scientific coding problems and verified Python solutions with unit tests for Turing's SciCode AI training set. 8-week contractor assignment, 40 hours a week, 12 countries. No rate published.

Mathematics graduates write chained scientific coding problems with verified Python solutions and unit tests for Turing's SciCode set. 8-week contractor assignment, 40 hours a week, 12 countries. No rate published.

Physics graduates author multi-part scientific coding problems with verified Python solutions and unit tests for Turing's SciCode AI training set. 8-week contractor assignment, 40 hours a week, open only to a list of 12 countries. No rate published.

Technical writers with some Python and JSON skills analyse public datasets, write structured documentation and explain reasoning in notebooks for Turing's research teams. 3 month contractor term, 20+ hours a week with PST overlap, open to ten named countries. No rate published.

Data analysts with 3+ years use Python and SQL to analyze ML training and evaluation outputs, define and validate metrics and diagnose failure modes for MLE Bench. Contract, 3 months, 20+ hours a week, ten named countries. No rate published.

ML engineers with 3+ years work on production-style ML codebases for MLE Bench benchmark tasks: training and inference pipelines, metrics, debugging. Contract, 3 months, 20+ hours a week, open to ten named countries. Live coding interview. No rate published.

Data engineers and data scientists build and validate pipelines, datasets and Python workflows for SWE Bench style AI evaluation. 3 month contractor term (adjustable), 20+ hours a week with 4 hours of PST overlap, open to ten named countries. Live coding interview. No rate published.

Prompt and policy specialists design and evaluate prompts for AI systems that use tools like Slack and PayPal, checking API and JSON parameters, SQL results, safety filtering and multilingual behaviour. Three-month contract, 20+ hours a week, open to ten named countries. No rate published.

Contract DevOps role at Turing: build data-validation steps into Docker-based pipelines, label datasets and schemas, and fail builds on bad data. 4+ years of DevOps, 2 to 4 weeks, 20 to 40 hrs/week, open to ten named countries.

C# engineers triage GitHub issues, set up Docker environments and check test quality to build verifiable software-engineering tasks that train and evaluate LLMs. Remote contract, 20-40 hours a week, open to nine named countries. No rate published.

Go engineers at tech-lead level pick open-source repositories, set them up in Docker and assess test quality so LLM bug-fixing can be measured. Contractor work at 20, 30 or 40 hours a week with 4 hours of PST overlap, for nine named countries. No rate published.

Ruby engineers at tech-lead level triage GitHub issues, set up repositories in Docker and check test quality to build LLM evaluation datasets. Remote contractor work at 20 to 40 hours a week with PST overlap, for nine named countries. No rate published.

Rust engineers triage GitHub issues, containerise repositories and judge test quality on a 3 month Turing contract building LLM evaluation data. 20 to 40 hours a week with PST overlap, open to nine named countries. No rate published.

Tech-lead-level C++ engineers triage GitHub issues, containerise repositories and judge test quality so LLMs can be evaluated on realistic bug fixes. Contractor assignment, 20 to 40 hours a week, 4 hours of PST overlap, open to nine named countries. No rate published.

Python engineers triage GitHub issues, containerise repositories and assess test quality for a 3 month Turing contract that builds LLM evaluation data. 20 to 40 hours a week, 4 hours of PST overlap, open to nine named countries. No rate published.