AI training jobs for software engineers

Remove a paper's core contribution code and package a coding task for AI agents to rebuild.

Remove a paper's core contribution code and package a coding task for AI agents to rebuild.

Remove a paper's core contribution code and package a coding task for AI agents to rebuild.

Package published computer architecture research code into tasks where an AI agent must rebuild the core contribution.

Package published GPU kernels research code into tasks where an AI agent must rebuild the core contribution.

Package published parallel algorithms and HPC research code into tasks where an AI agent must rebuild the core contribution.

Package published networked systems research code into tasks where an AI agent must rebuild the core contribution.

Package published programming languages research code into tasks where an AI agent must rebuild the core contribution.

Package published robotics and reinforcement learning research code into tasks where an AI agent must rebuild the core contribution.

Build coding tasks from published security research by removing code and defining held-out evaluation workloads.

Build coding tasks from published software engineering research by removing code and defining held-out evaluation workloads.

Write programming and system design tasks with solutions, and evaluate AI-generated code and explanations for correctness, security and maintainability.

Design realistic software engineering tasks and assess AI-generated code, writing reference solutions, tests, and criteria.

Review real engineering tasks and judge the quality of the work, as you would review pull requests.

Review and rank technical slide decks from your field to determine what makes an excellent technical presentation.

Solve and validate Rust backend problems involving service behavior, APIs, persistence, concurrency, performance, and reliability.

Submit web app and website designs you created for review as part of a licensing program.

Create, evaluate, and refine AI-generated documents, spreadsheets, and slide decks for cybersecurity compliance workflows like security risk assessment.

Create, evaluate, and refine AI-generated technical design documents, data pipeline documentation, and API specification guides.

Evaluate and refine AI-generated blameless postmortems, root-cause analyses, on-call runbooks, and SLO and error-budget reports.

Evaluate and refine AI-generated security risk assessment and control matrices to help train a language model.

Create, evaluate, and refine AI-generated technical design documents and API specification guides for language model training.

Evaluate and refine AI-generated blameless postmortems and on-call runbooks for language model training.

Solve self-contained coding, debugging, and technical reasoning challenges by selecting appropriate algorithms and data structures and producing reproducible solutions.

Build reproducible reinforcement learning environments testing AI models on fixing code, adding features, and refactoring, with golden reference solutions.

Build reproducible reinforcement learning environments testing whether AI models can fix code, with a golden reference solution.

Conduct detailed code reviews and contribute real-world programming examples and explanations as input for AI training datasets.

Build small game projects in a major game engine with realistic gameplay bugs, and write reference solutions for each task.

Build reproducible mobile coding task environments and write deterministic verifiers that accept correct solutions and reject incorrect ones.

Create challenging coding tasks involving feature implementation and bug fixes, and implement deterministic verifiers that accept valid solutions.

Create realistic coding tasks for AI involving feature implementation and bug fixes, and implement deterministic verifiers.

Review, build, and verify realistic coding challenges and evaluation harnesses for testing AI agents.

Review and verify realistic programming tasks and evaluation harnesses used to evaluate AI agents.

Build realistic software engineering tasks and judge frontier AI model answers against professional standards.

Evaluate LLM-generated code patches for repositories you maintain and provide detailed written feedback and justification for each assessment.

Create, evaluate, and refine AI-generated work across core software engineering workflows for language model training.

Create, evaluate, and refine AI-generated work across core software engineering workflows to help train a language model.

Create, evaluate, and refine AI-generated work across core software engineering workflows for language model training.

Create, evaluate, and refine AI-generated work across core software engineering workflows to help train a language model.

Create challenging coding tasks for AI involving feature implementation and bug fixes, plus deterministic verifiers.

Create realistic coding tasks involving feature implementation and bug fixes for AI to solve, and implement deterministic verifiers.

Review and verify programming tasks used for testing AI agents, evaluating coding harnesses and the quality of realistic coding challenges.

Explore unfamiliar software repositories, debug front-end and back-end components, and review AI-generated patches for correctness, completeness, security, and maintainability.

Develop and debug front-end and back-end application components, and review AI-generated solutions for correctness, performance, security, and maintainability.

Create custom competitive programming problems, reference solutions, and robust test suites for reinforcement learning environments that test AI models.

Write engineering problems with reference answers, and grade model-written code and design for edge cases, test quality, and production failures.

Design and validate research-grade machine learning benchmark tasks that measure what frontier AI models can do.

Develop and refine new features within existing open source codebases and analyze and resolve complex bugs.

Design and build scalable big data pipelines, and implement data integration and transformation using Python.

Debug and resolve issues across codebases, and develop and review real-world coding tasks to evaluate AI model performance.

Review coding agents' generated code in real-world repositories, determine technical correctness, and identify failure modes.

Create reinforcement learning environments and golden reference solutions testing AI models on fixing code and creating features.

Discuss CI/CD pipelines and security practices in an interview study on application features and vulnerability verification.

Find real vulnerabilities in private repositories and turn them into difficulty-probed cybersecurity tasks for frontier LLMs.

Author cloud scenarios from architecture design through incident response and grade model output for correctness, security, and cost efficiency.

Author security scenarios from alert triage through incident response and grade model output for technical accuracy.

Author realistic agile delivery scenarios and grade model output for agile soundness and team dynamics awareness.

Design realistic technical problems, author expert-level reference solutions, and evaluate submitted work for correctness and depth.

Evaluate AI-generated analyses of cybersecurity scenarios and exploits, and review technical writeups for correctness and mitigation quality.

Evaluate how frontier AI models reason about hard, expert-level problems in ML inference and GPU kernel engineering.

Complete software development, debugging, code review, and technical implementation tasks across a variety of stacks and applications.

Review AI-generated SQL queries and database explanations, and provide precise written feedback against project guidelines.

Review AI-generated Python code and trainer work, then provide precise written feedback against project guidelines.

Review AI-generated HTML/CSS code and trainer work, evaluating quality against project guidelines and giving precise written feedback to contributors.

Review AI-generated C code and trainer/QA work, evaluating quality against project guidelines and providing precise written feedback.

Review AI-generated C# code and trainer work against project guidelines, providing precise written feedback on quality issues.

Review AI-generated C++ code and trainer work against project guidelines, providing precise written feedback.

Review AI-generated Java code and trainer/QA work against project guidelines, providing precise written feedback.

Review AI-generated Go code for correctness and concurrency safety, and give written feedback to trainers following project guidelines.

Review AI-generated PHP code and trainer work against project guidelines, and give precise written feedback on correctness, security, and quality.

Review AI-generated TypeScript code and trainer work against project guidelines, providing precise written feedback.

Review AI-generated JavaScript code against project guidelines and give written feedback to trainers and QAs on correctness and quality.

Review AI-generated Kotlin code and trainer/QA work, providing precise written feedback against project guidelines.

Review AI-generated Dart code and trainer work, evaluating quality against project guidelines with precise written feedback.

Review AI-generated R code and trainer work, evaluating statistical validity and reproducibility against project guidelines and giving precise written feedback.

Review AI-generated Bash code and trainer work against project guidelines, providing written feedback on correctness and safety.

Review AI-generated iOS code and technical explanations, and rate and compare multiple AI responses on correctness and reasoning quality.

Review AI-generated Android code and technical explanations, and write reference implementations with step-by-step reasoning.

Review AI-generated React Native code and mobile app architectures, and write reference implementations with step-by-step reasoning.

Review AI-generated full-stack code and system designs, and produce reference implementations with step-by-step reasoning.

Review AI-generated Angular code and rate and compare multiple AI responses based on correctness and reasoning quality.

Review AI-generated React/Next.js code and technical explanations for correctness, and write reference implementations demonstrating best practices with step-by-step reasoning.

Review AI-generated SQL queries and database designs, and write reference solutions with step-by-step explanations for complex data problems.

Generate code repositories containing OSS-fuzz-style vulnerabilities and produce correct patches for them.

Review AI-generated Scala code and technical explanations, and write reference implementations with step-by-step reasoning for complex engineering problems.

Review AI-generated C++ code and technical explanations for errors in logic, memory handling, and concurrency, and write reference model solutions.

Review AI-generated C code for errors in logic, memory handling, and concurrency, and write reference implementations.

Review AI-generated PHP code and backend designs, then rate and compare AI responses for correctness.

Review AI-generated Swift and iOS responses, evaluating their reasoning quality, and write expert-level explanations and model solutions.

Review AI-generated Ruby and Rails code, evaluate reasoning quality, and rate and compare multiple AI responses for correctness.

Review AI-generated responses and write R data-analysis solutions, evaluating the reasoning quality behind each solution.

Review AI-generated Kotlin responses and rate and compare multiple AI responses for correctness and reasoning quality.

Review AI-generated C#/.NET responses, fact-check technical claims, and rate and compare responses for correctness and reasoning quality.

Review AI-generated Rust code and explanations, evaluate reasoning quality, and rate and compare multiple AI responses for correctness.

Review AI-generated JavaScript code and explanations, and rate and compare multiple AI responses on correctness and reasoning quality.

Review AI-generated Go code and compare multiple AI responses for correctness and reasoning quality.

Review AI-generated TypeScript code snippets and explanations for correctness, and write reference solutions while rating and comparing multiple AI responses.

Review AI-generated Java code and technical explanations, then rate and compare multiple AI responses for correctness.

Review AI-generated Python solutions for correctness and code quality, and write clear step-by-step explanations and exemplary solutions.

Complete software development, debugging, code review, and technical implementation tasks across a variety of stacks and applications.

Design and evaluate iOS development scenarios that test an AI's ability to apply Swift programming principles.

Design and evaluate coding scenarios that test how AI interprets programming principles across software development, debugging, and system design.

Train and evaluate AI models in Frontend Engineering and create tasks and deliverables based on real-world scenarios.

Train and evaluate AI models in Full-Stack Engineering, and create tasks and deliverables based on real-world scenarios.

Build end-to-end applications across various domains that become part of an AI training dataset.

Build complete, functional end-to-end applications across various domains and use cases that become AI training examples.

Turn completed pull requests into engineering tasks, and build evaluations using automated tests, shell commands, and LLM grading prompts.