Aidapt is an embedded AI transformation consultancy. We don't sell software — we embed into MEA businesses and build the operational infrastructure they need to compete in an AI-first world.
We're hiring a junior Agentic AI Engineer who thinks in agents, not scripts.
You won't inherit a codebase to maintain. You'll be handed a real client problem and asked to figure out how AI solves it — then build that solution. That's the whole job.
This is a pure build role. You'll design and ship agentic workflows, LLM-powered automations, and AI integrations that plug directly into our clients' operational stacks. If you've already built something with Claude, n8n, or LangChain — even a side project — you're exactly who we're looking for.
What you'll build:
- AI agents and agentic workflows on LangChain / LangGraph and the Anthropic Claude API — multi-step reasoning, tool use, memory, structured outputs
- End-to-end automation pipelines on n8n and Make.com — webhook triggers, branching logic, parallel flows, error handling, retries
- ClickUp integrations — task creation, custom field handling, list routing, pipeline automation via the ClickUp API
- WhatsApp-native workflows using WATI — lead-capture bots, automated follow-ups, customer-journey automation over WhatsApp Business
- Prompt engineering pipelines — system prompts, multi-turn conversation design, tool-calling chains, hallucination mitigation
- Internal AI tooling for Aidapt's own delivery Pods
Requirements
Required
- Hands-on experience (personal projects count) with LangChain or LangGraph building agent pipelines
- Working knowledge of the Anthropic Claude API — tool use, system prompts, streaming, structured output patterns
- At least one n8n or Make.com workflow built end-to-end (not just watched a tutorial)
- Comfortable making raw API calls — REST, webhooks, JSON payloads, authentication patterns
- Can read and write clean Python or JavaScript
- A problem-first thinker — you understand what you're solving before you pick the tool
Nice to have
- ClickUp API experience — custom field UUIDs, task relationships, list operations
- WATI or any WhatsApp Business API experience
- Exposure to other LLM providers — OpenAI, Gemini, or similar
- RAG pipelines, vector databases, or embeddings (basic understanding is enough)
- Awareness of prompt injection risks, AI safety basics, and responsible deployment
Who this is NOT for
- People who want to train models from scratch — we deploy AI, we don't build foundational models
- People who need a detailed ticket queue to know what to work on
- People who Google "what is LangChain" before sending in their application
Benefits
Why Aidapt
- You'll ship real AI systems into real businesses across MEA — not demos that die in a slide deck
- You'll work inside a small, senior engineering Pod — no layers, no bureaucracy, no waiting for approvals
- You'll be part of Aidapt's AI architecture decisions from day one
- The MEA AI market is at an inflection point. We're building the infrastructure layer for it — and you'll have a front-row seat
Applications without a build sample will not be reviewed.
Aidapt is an equal opportunity employer. We evaluate candidates on what they've built, not where they studied.