Devops & Backend Specialist

Role: AI Training Data Engineer (DevOps & Backend Specialist) Type: Freelance / Contract | Remote (India) Experience: 3–8 years About the Work: You will be contributing to the development of next-generation AI systems by creating and curating high-quality training data for reinforcement learning of agentic AI models. This includes writing realistic engineering tasks, generating and evaluating code solutions, and providing expert judgment on AI-generated outputs. Your work directly shapes how AI agents learn and improve. What You'll Do: - Design and document realistic DevOps and backend engineering scenarios that serve as benchmark tasks for AI training - Write, review, and rank code solutions (Python, Bash, infrastructure-as-code) that an AI agent would be evaluated against - Evaluate AI-generated code and agentic task completions for correctness, security, efficiency, and style - Create detailed acceptance criteria and ground-truth solutions for complex engineering workflows - Contribute to CI/CD, observability, Kubernetes, and infrastructure-as-code task libraries - Collaborate with the team to ensure task diversity, difficulty calibration, and real-world relevance Required Skills & Experience: - 3–8 years of hands-on software engineering experience (product companies strongly preferred) - CS degree at undergraduate level (must have) - Strong proficiency in Python and at least one of: Bash, Go, or similar - Solid experience with Kubernetes, Docker, and container orchestration - Hands-on experience with CI/CD pipelines (GitLab CI or GitHub Actions preferred) - Working knowledge of Terraform or other infrastructure-as-code tools - Familiarity with at least 3 of the following: Prometheus + Grafana, Keycloak/IAM, gRPC, k3s/k9s, PostgreSQL, Redis, API design

 

Strongly Preferred: - Experience writing technical documentation, runbooks, or engineering specs - Ability to clearly articulate why a solution is correct, efficient, or secure - Exposure to LLM-based coding tools (e.g., GitHub Copilot, Cursor, Claude Code) — understanding their strengths and failure modes is a plus - Prior experience in AI data annotation, RLHF, or technical content creation is a significant advantage Work Dynamics: - Fully remote, freelance engagement from anywhere in India - Output-based compensation tied to quality, completeness, and technical complexity of deliverables - Clear acceptance criteria provided upfront for every task - Flexible pacing — high-performing contributors typically complete multiple tasks per week Must have a high-performance laptop/desktop and reliable high-speed internet