Software Engineer (C++ - LLM Evaluation & Repository Validation)

Are you passionate about open-source software and large language models (LLMs)? We're looking for a talented Software Engineer (C++) to join our research-driven project focused on evaluating and training LLMs using real-world GitHub repositories.
In this role, you’ll dive deep into high-quality public C++ repositories (5k+ stars), triage issues, assess test coverage, and create verifiable software engineering tasks that challenge current-generation LLMs. You’ll collaborate with researchers, and have the opportunity to lead junior engineers on cutting-edge AI training datasets.🔧 What You’ll Be DoingAnalyze, triage, and evaluate real GitHub issues across popular open-source C++ projectsSet up dev environments, Dockerize codebases, and ensure seamless local runsAssess unit test coverage, code quality, and debugging scenariosModify and run repositories locally to benchmark LLMs in bug-fixing and real-world code understandingWork closely with AI researchers to select, structure, and validate SWE tasksOption to mentor or lead junior contributors.✅ What You Bring3+ years of strong hands-on experience with C++Proficiency in Git, Docker, and navigating complex GitHub repositoriesExperience with development environment setup and pipeline configurationComfortable with evaluating test coverage and debugging open-source codebasesBonus if you've worked on or contributed to open-source repos (5k+ stars)Extra credit for involvement in LLM projects, developer tools, or automation.💡 About the ProjectWe're creating robust evaluation and training datasets for LLMs by modeling real developer workflows—bug triaging, environment setup, testing, and code modification—based on the history of well-known GitHub repositories.Skills: problem solving,ai training datasets,code quality,c++,git,open-source software,github,docker,debugging,unit test coverage