[Hiring] MCP Expert @24-MAG

Role Description

We are sharing a specialised part-time consulting opportunity for experienced software engineers with strong expertise in Python, Java, Rust, C++, Go, TypeScript, algorithms, debugging, refactoring, and performance optimisation to contribute to an advanced AI training project involving Model Context Protocol (MCP) environments. Selected professionals will create reinforcement-learning environments that test an AI model's ability to solve complex software-engineering problems using MCP tools and real server interactions. The work combines practical software engineering, deterministic evaluation design, and the creation of high-quality reference solutions. No prior experience in AI is required.

Key Responsibilities

• MCP Environment Development
  • Create reinforcement-learning environments based on realistic software-engineering tasks
  • Design scenarios requiring agents to discover and reason over information from MCP servers
  • Build tasks involving real tool interactions rather than isolated code-generation exercises
  • Ensure environments accurately measure both MCP tool use and engineering capability
  • Maintain reproducibility across evaluation runs
• Software Engineering Task Design
  • Create challenging tasks involving bug fixing, feature implementation, refactoring, and performance optimisation
  • Develop scenarios that require meaningful reasoning across existing codebases
  • Design tasks that test algorithms, data structures, debugging, and architectural judgement
  • Ensure problems reflect realistic engineering constraints and workflows
  • Balance task complexity with clear, measurable success criteria
• Golden Solutions & Deterministic Verification
  • Create high-quality golden reference solutions for evaluation tasks
  • Develop deterministic verification logic that reliably distinguishes correct from incorrect implementations
  • Define clear acceptance criteria for software behaviour and task completion
  • Validate environments against edge cases and unintended solution paths
  • Ensure evaluation logic remains stable and reproducible
• Code Quality & Performance Engineering
  • Debug complex software issues across multiple programming languages
  • Implement maintainable features in existing codebases
  • Refactor code while preserving intended functionality
  • Identify and resolve performance bottlenecks
  • Apply scalability, maintainability, and software-quality best practices
• Technical Review & Collaboration
  • Review task quality, code correctness, and evaluation robustness
  • Communicate technical decisions and assumptions clearly
  • Participate in collaborative review of software-engineering environments
  • Contribute to code-review standards and engineering best practices
  • Work effectively in remote and cross-functional technical teams

Qualifications

• Strong proficiency in one or more of C++, Python, Java, Go, TypeScript, or Rust
• Deep understanding of algorithms, data structures, and performance optimisation
• Demonstrated experience debugging complex software issues
• Strong background in feature development and codebase refactoring
• Proven ability to improve software performance and scalability
• Experience working with large or distributed codebases is highly valuable
• Familiarity with rigorous code-review practices and software-engineering standards
• Strong written and verbal communication skills
• High attention to technical detail and reproducibility
• Experience with modern AI or machine-learning systems is beneficial but not required
• Prior AI-training or model-evaluation experience is not required

Engagement Details

• Part-time independent contractor engagement
• Fully remote
• Compensation: $60–$120/hour
• Expected commitment: approximately 15 hours per week
• Schedule is flexible, including the option to work evenings or weekends
• Compensation is output-based, with payment made for tasks that meet project specifications
• Minimum weekly submission requirements apply
• Work will involve MCP-based reinforcement-learning environments, software-engineering task design, deterministic verification, and golden reference solutions
• The selection process may include screening questions, an approximately 30-minute AI interview, a technical assessment, and hiring-manager review
• Selected professionals should be prepared to begin their first tasks within approximately 24–48 hours of completing onboarding
• Roles are typically filled within approximately 48 hours
• Work must be completed without using confidential or proprietary information belonging to any employer, client, institution, or other third party