Senior Software Engineer – C++ (LLM Evaluation & Repository Validation)
Overview
Turing is assembling a team to build LLM evaluation and training datasets that teach models to solve real software engineering tasks. This project constructs verifiable SWE problems by mining public repository histories and using a synthetic, human-in-the-loop approach. A senior C++ engineer in this role will analyze trending GitHub issues, triage bugs, and evaluate unit test quality across open-source libraries. The work includes setting up repositories with Docker, running and modifying codebases locally, and partnering with researchers to design challenges that stretch LLM capabilities. This is a fully remote contractor assignment with required PST overlap.
What You'll Do7
- 1Analyze and triage GitHub issues across popular open-source libraries to identify strong LLM challenge candidates.
- 2Set up and configure code repositories, including Dockerization and local environment automation.
- 3Evaluate unit test coverage and quality to determine whether tasks are verifiable.
- 4Run and modify codebases locally to test LLM performance on bug-fixing scenarios.
- 5Collaborate with researchers to select repositories and issues that are difficult for LLMs.
- 6Expand dataset coverage by contributing tasks across different programming languages and difficulty levels.
- 7Lead or mentor a team of junior engineers on related projects when needed.
Requirements7
- 1Minimum 3 years of professional software engineering experience.
- 2Strong proficiency in C++.
- 3Hands-on experience with Git, Docker, and basic software pipeline setup.
- 4Ability to understand and navigate complex, unfamiliar codebases.
- 5Comfort running, modifying, and testing real-world projects locally.
- 6Experience contributing to or evaluating open-source projects, which is a plus.
- 7Prior involvement in LLM research or evaluation projects, which is a nice to have.
Who Should Apply
The ideal candidate for this role is a senior C++ engineer who reads open-source code daily, can Dockerize a repo in minutes, and judges test suites by their ability to catch regressions. This person enjoys the detective work of triaging GitHub issues and has a feel for which bugs will stump an LLM. The role is less suitable for developers who want to build new products from scratch or who stay inside their own codebase. Candidates often get rejected when the technical interview reveals shallow Git and build-pipeline skills, or when they cannot commit to the required 4-hour daily overlap with PST. Another common failure is assuming LLM experience matters more than solid C++ and repository engineering fundamentals.
Location
Required Skills
Application Tip
Practice a 60-minute technical interview that mirrors the evaluation: pick a public C++ repository, set it up with Docker, identify a real bug from its issue tracker, and walk through how you would verify a fix and measure test coverage. Prepare one quantified example of a complex codebase you navigated and changed.
See NearSkill jobs more often in your search
How your application is processed
1Application received
Your resume and details are logged the moment you apply.
2ATS + eligibility screening
We check your profile against the role’s skills, seniority, and requirements.
3Employer sees qualified profiles only
Only candidates who clear screening move forward.
Similar open positions
Explore active roles that match your skills and interests.
Turing
VerifiedSenior Software Engineer – C#(LLM Evaluation & Repository Validation)
This project builds LLM evaluation and training datasets that help models solve realistic software engineering problems. The work involves creating verifiable software engineering tasks from public repository histories using a synthetic, human-in-the-loop approach. Engineers analyze GitHub issues, set up Docker-based environments, and evaluate test coverage to judge model performance in bug-fixing scenarios. The role fits a tech lead-level engineer comfortable with C#, Git, and running complex codebases locally.
Turing
VerifiedSenior Software Engineer – LLM Evaluation & Repository Validation
This contract role centers on building LLM evaluation datasets from public repository histories. You will triage GitHub issues, set up Docker environments, and run codebases on your own machine to score how well language models fix real bugs. The work targets popular repositories with 500+ stars and spans languages such as Python, JavaScript, and Go. The project uses a human-in-the-loop approach to expand task coverage across difficulty levels and programming languages.
Turing
VerifiedSenior Software Engineer – Python (LLM Evaluation & Repository Validation)
GitHub repository histories become training ground for LLM evaluation at this contract role. You will create verifiable software engineering tasks through a synthetic, human-in-the-loop pipeline that broadens dataset coverage across programming languages and difficulty levels. Day-to-day work involves triaging issues in trending open-source libraries, configuring Docker environments, and measuring unit test quality. You will run and modify local codebases to see how well LLMs handle real bug-fixing scenarios, then share findings with researchers. There is also room to take on a team lead role with junior engineers.
Turing
VerifiedSenior Software Engineer – Ruby (LLM Evaluation & Repository Validation)
The project builds LLM evaluation datasets and verifiable software engineering tasks drawn from public repository histories. Ruby engineers at tech-lead level will triage GitHub issues, set up repositories with Docker, and run codebases locally to score model performance. The day-to-day includes test coverage review, environment automation, and collaboration with researchers to pick challenging bug-fixing problems. The role also offers the chance to lead a small team of junior engineers while working on advanced AI projects.
Turing
VerifiedSenior Software Engineer – Rust (LLM Evaluation & Repository Validation)
This role sits inside a project that produces LLM evaluation and training datasets for real-world software engineering problems. The team builds verifiable SWE tasks from public repository histories using a synthetic approach with human-in-the-loop review, then expands coverage across programming languages and difficulty levels. You will analyze and triage GitHub issues in trending open-source libraries, configure repos with Docker, run codebases locally, and assess how well LLMs handle bug-fixing scenarios. The work is hands-on and spans environment automation, test coverage evaluation, and collaboration with researchers to select challenging repositories and issues. You also have the option to lead a small team of junior engineers.
Turing
VerifiedSenior Software Engineer – Go (LLM Evaluation & Repository Validation)
This role centers on building LLM evaluation datasets that teach models to solve realistic software engineering problems. You will triage GitHub issues from popular open-source repositories, set up Docker containers and code environments, and run modified codebases in local environments to measure how well models handle bug-fixing tasks. The work is hands-on and includes evaluating unit test coverage, collaborating with researchers to identify challenging tasks, and the option to lead junior engineers. Projects draw on public repository histories with a human-in-the-loop, synthetic approach to expand task coverage across Go and other programming languages.