Create realistic, challenging coding tasks for frontier AI models from real repositories.
Build reproducible environments and write deterministic verifiers for correct solutions.
Provide reference solutions and flag ambiguous or unverifiable tasks.
Google is a technology company that develops products and services to organize the world's information. They are a large global employer with a culture focused on innovation and engineering excellence.
Work independently and asynchronously while maintaining high standards of technical quality.
Our client is an innovative technology organization developing advanced artificial intelligence systems through real-world software engineering challenges. CONFISA Group is supporting the recruitment of qualified professionals for this fully remote, project-based opportunity, with a focus on experienced software engineers.
Design realistic software engineering challenges to assess AI models' coding abilities.
Build reference solutions and automated verification tests to ensure correctness.
Work independently on a fully remote, contract basis with flexible scheduling.
Jobgether is an AI-powered job matching platform that connects candidates with hiring companies through objective, AI-driven reviews. The company focuses on efficient hiring and shares top-fitting candidates directly with employers, with a culture emphasizing data privacy and fair evaluation.
Evaluate AI-generated coding interactions from beginning to end, assessing quality and effectiveness.
Provide detailed, opinionated written feedback on AI outputs and reasoning.
Help define and refine quality benchmarks for AI coding agents.
The company focuses on improving AI coding agents through expert evaluation. It operates with a small, remote team and values technical excellence and critical thinking.
Work on your personal software project using an AI coding assistant to build features, fix bugs, refactor code, and improve documentation.
Review and compare two versions of AI-generated coding responses, selecting the more useful and accurate one.
Provide clear, honest feedback explaining your choice, without using AI assistance for evaluations.
An enterprise client is seeking experienced software developers for an AI training project. The project aims to improve AI coding tools by gathering feedback from developers working on their own projects.
Analyze, interpret, and debug code snippets and technical documentation across Python, JavaScript, C++, Java, and SQL.
Produce comprehensive explanations or annotations for code segments, focusing on clarity and accuracy.
Review, assess, and improve solutions generated by AI relating to software development challenges.
The company engages coding experts to contribute to customer AI training projects by analyzing and debugging code across multiple languages. The company operates remotely, emphasizing independent work and high-quality human input to improve AI capabilities.
Develop, debug, and optimize front-end and back-end components for AI training projects.
Create, solve, review, and validate complex software engineering tasks across the full application stack.
Translate software implementations between programming languages and evaluate AI-generated code for correctness and quality.
The client is a technology organization working on advanced artificial intelligence initiatives that leverage real-world software engineering expertise. It is a fully remote, global organization of undisclosed size, emphasizing independent contributions and high-quality engineering with flexible schedules and output-based compensation.
Debug and resolve issues across complex, multi-file codebases.
Implement new backend or full-stack functionality and refactor existing systems.
Develop and extend unit and integration test suites and review code for correctness.
Anyone AI is an AI lab recruiting experienced Python Software Engineers for paid, project-based engagements. They are a flexible consulting opportunity seeking engineers with strong technical skills for diverse software projects.
Judge the AI's work by reviewing its output like a teammate's pull request.
Pair with the AI on your own task to flag where the model goes wrong.
Work on your own schedule with a fixed-pay contract earning $6,000/wk for 40hrs.
Handshake powers a platform connecting 25 million job seekers, 1 million+ employers, and 1,600 educational institutions, and has built a fast-growing AI data business working with frontier AI labs. They have grown to a ~$1B run rate, paying ~$60M to over 30,000 individuals monthly, with a culture focused on shaping careers in the AI economy.
Direct AI agents on complex development tasks and critically evaluate their output.
Write high-quality code in languages such as Python, Java, C++, C#, or Ruby.
Collaborate with a high-caliber team in a remote atmosphere with minimal travel.
We are an innovative and rapidly growing software company providing SaaS applications for tax collection, electronic payments, and bond auctions. We reward teamwork, professional excellence, and individual responsibility, and offer a technology-rich work environment with remote and on-site flexibility.
Build and maintain features across CLI, backend, and frontend with AI coding agents as a core part of your workflow.
Take ownership of projects from idea through deployment and beyond, making architecture and design decisions.
Collaborate closely with teammates, operationalize quality, and contribute to a culture of support and excitement.
Minimal builds a secure, reproducible foundation for software with private artifact registries and policy enforcement. They are a small, remote team that values predictable and effortless software development.
Apply advanced mechanical engineering expertise to create, solve, and validate engineering tasks for AI training.
Use computational modeling, simulation, and AI coding agents to develop and assess engineering solutions.
Work fully remotely as an independent contractor on a 12-month contract, contributing to AI system reliability.
We are an AI-powered job matching platform that connects top candidates with hiring companies. We use AI to review applications and shortlist the best-fitting candidates for a fair and efficient recruitment process.
Design, build, and deploy AI agents that automate real business processes.
Integrate LLMs from OpenAI, Gemini, Anthropic, and open-source models into products.
Own delivery end-to-end, from prototype to production, choosing the right LLM/tooling.
We are a software development agency based in Vilnius, Lithuania, building production-ready AI and software solutions. Our team works remotely across the EU/EEA, with a focus on independent, fast-moving delivery.
Build AI-first SaaS products, including LLM-powered workflows, retrieval over compliance content, and in-product agents.
Develop and optimize agentic, pipeline-driven development systems where AI coding agents plan, write, test, and review code under strong engineering oversight.
Maintain high engineering standards with secure, tested, scalable solutions and participate in on-call rotation.
Mitratech builds world-class software products that simplify operations in Legal, Risk, Compliance, and HR functions. With over 35 years of expertise, the company serves 20,000 client companies globally, including 30% of the Fortune 500, and fosters a diverse, inclusive culture that blends entrepreneurial spirit with enterprise investment.
Evaluate AI-generated Java solutions for correctness, efficiency, and maintainability.
Analyze object-oriented design, concurrency, JVM behavior, data structures, algorithms, and architecture.
Document recurring model failure patterns and communicate findings in a structured, actionable way.
Jobgether is an AI-powered job matching platform that connects candidates with hiring companies. It posts roles on behalf of partner employers and emphasizes objective, fair application reviews.