GenAI Engineer — LLM Agents & AI Automation

Clinic Robo is an AI-powered revenue and operations platform for aesthetic and dermatology clinics — live in real clinics today. The product runs a multi-agent AI architecture: AI sales agents that qualify and follow up leads over WhatsApp, voice-to-note transcription and clinical scribing, automated follow-up sequencing, lead scoring, and AI-generated reporting — all wired into a production CRM with real revenue flowing through it. We are hiring a GenAI engineer who builds with AI, not just around it. You should be the kind of developer who ships daily using Claude Code, Cursor, or similar AI-first tooling, and who has actually taken LLM agents to production — handled hallucinations, structured outputs, tool calling, cost, and latency — not just built a chatbot demo. You will work directly with the founder and own the AI layer of the product. **What you will build** * Multi-agent workflows: AI SDR agents that read lead context from Postgres, decide next actions, and execute them (WhatsApp messages, task creation, escalation to humans) * Tool-calling pipelines: connecting Claude/GPT to live CRM data with function calling, structured outputs (JSON schemas), and guardrails so agents never act on bad data * Voice AI: speech-to-text pipelines, call transcription and summarization, voice-note-to-clinical-note workflows including translation * Prompt systems: versioned, testable prompt architectures — with evals so we know when a change makes agents better or worse * AI content generation: automated ad copy, follow-up messages, and report narratives grounded in real clinic data * The plumbing that makes it reliable: retries, fallbacks between model providers, token/cost monitoring, and latency optimization **Our stack** * AI: Anthropic (Claude) and OpenAI APIs — function calling, structured outputs, streaming; STT/voice pipelines * Dev workflow: AI-first — Claude Code / Cursor daily; we care about output speed and code quality, not hours typed * Backend: Supabase (PostgreSQL, RLS, Edge Functions, Realtime), webhook-driven integrations * Frontend: React 18, TypeScript, Vite, Tailwind; Expo (React Native) for mobile/web * Integrations: Meta Lead Ads webhooks, WhatsApp Business API platforms, call-recording APIs; deployed on Vercel **What we are looking for** * 2+ years building production software, with at least 1 year shipping LLM-powered features to real users (not side projects only — show us something people actually used) * Deep hands-on experience with Claude and/or OpenAI APIs: tool use / function calling, structured outputs, prompt design, context management * Daily driver of AI coding tools (Claude Code, Cursor, Copilot agents) — you can describe your workflow and how you verify AI-written code * Solid TypeScript and SQL fundamentals — agents are only as good as the data layer they sit on * Pragmatic about reliability: you think in terms of evals, fallbacks, idempotency, and what happens when the model is wrong * Good English; based in Dubai or able to relocate (on-site role) **Nice to have** * Multi-agent frameworks or hand-rolled agent orchestration in production * RAG pipelines, embeddings, and vector search (pgvector a plus) * Voice AI experience: Whisper/Deepgram-style STT, diarization, or real-time transcription * WhatsApp Business API, Meta Graph API, or CRM/health-tech experience * MCP (Model Context Protocol) servers or building tools for AI agents **Why join us** * Your agents run in live clinics from week one — real leads, real bookings, real revenue, immediate feedback * AI-first engineering culture: small team, no bureaucracy, direct line to the founder * Freedom to pick the best model/tool for the job — we pay for the good stuff * Growth path to lead the AI engineering function as the platform scales across clinics and countries **How to apply** send the email at [email protected] Pay: AED2,509.44 - AED9,526.72 per month Work Location: In person