You’ll run our AI interview pipeline end-to-end: design and maintain prompts, set up interview flows in our system, QA model outputs, and keep everything consistent across roles, languages, and clients . You’re detail-obsessed, comfortable with structured text, and can think like a recruiter and an engineer.
Responsibilities:
– Curate and version prompt packs for roles (JD → competencies → questions → scoring).
– Configure interview flows in our Django admin (nodes, timers, hints, language, proctoring flags).
– Create rubrics & scoring guides (behavioral, KPI-based, red-flags).
– Run A/B tests across models (OpenAI/Groq/Anthropic/Gemini) and log results.
– Do multilingual consistency checks (en → hi → en back-translations). QA transcripts, flags (cheating, silence), and edge-case triage.
– Maintain a Prompt Library (versioned, tagged, reviewed). Produce weekly ops reports (completion %, pass rates, anomalies, time-to-offer).
Must-Have Skills:
– Strong written English Understands LLM prompt design (system/user, few-shot, delimiters, tool use).
