Full-Stack Developer - AI Trainer

About Anyone AI: Anyone AI is a forward-thinking artificial intelligence talent ecosystem that connects elite, highly skilled software engineers with prominent, market-leading AI Labs and frontier research institutions. We focus on accelerating the development of next-generation foundational models, specializing in advanced source code alignment, RLHF verification layers, and high-fidelity algorithmic training. By building robust evaluative pipelines, we help shape the future of autonomous software engineering tools at scale.

Position Overview

We are seeking highly skilled, polyglot Full-Stack Developers to step into an expert project-based consulting role as an AI Trainer, supporting a leading global AI Lab. In this specialized track, you will not build traditional user-facing apps; instead, your codebase will serve as the benchmark training grounds for advanced LLMs. You will design, implement, and audit highly realistic multi-file coding tasks spanning bug resolution, legacy migrations, and test harness generation. This role demands a high level of code literacy, a strong sense of edge-case failure modes, and the ability to author crystal-clear natural-language software specifications.

Key Responsibilities

• AI Task Architecture: Design and implement complex, production-realistic multi-file programming scenarios covering bug fixes, custom features, and repository refactoring.
• Specification & Reference Coding: Write clean, well-documented reference implementations paired with clear, explicit natural-language technical specifications to train model alignment hooks.
• Test Suite Engineering: Develop and extend comprehensive unit and integration verification suites across polyglot execution environments using frameworks like pytest, Jest, JUnit, or Go testing.
• Peer Review & Quality Control: Review task submissions generated by peer consultants, evaluating them rigorously for technical correctness, structural clarity, and real-world enterprise realism.
• Edge-Case Analysis: Proactively identify runtime ambiguities, system vulnerabilities, and unexpected code failure modes within generated coding setups to maintain high-signal training datasets.

Required Skills & Qualifications

• 3–7 years of verified professional history running backend or full-stack software development in live production systems.
• Strong, high-level engineering command over **Python and JavaScript/TypeScript**, alongside a functional working knowledge of **Java, C#, or Go**.
• Proven background navigating, debugging, and refactoring large, sophisticated multi-file codebases.
• Practical execution history with testing abstractions (e.g., pytest, Jest, JUnit, xUnit, or Go testing frameworks).
• Advanced, highly professional written and verbal English communication proficiency.
• Location Context: 100% remote-first project framework open exclusively to qualified software engineers permanently residing in India.
• Engagement Logistics: Part-time, project-based consulting structure allowing complete schedule control, where estimated completions calibrate the hourly rewards tier.

Preferred Strategic Indicators (Nice to Have)

• Prior experience participating in code generation indexing, developer LLM benchmarking, or prompt engineering workflows.
• Background executing massive codebase migrations or structural legacy architecture refactoring.

What We Offer

• Targeted Compensation Range: $15 – $35 USD per hour (Paid on a per-project basis, estimated transparently based on anticipated task completion intervals).
• The exceptional opportunity to collaborate directly with a world-leading AI Lab, defining how foundation models interpret and write code.
• Absolute scheduling autonomy, allowing you to easily maintain existing external vendor or employment commitments.
• A pure, intellectually stimulating remote-first ecosystem optimized for engineering experts who love deep code exploration and code review dynamics.