AI Engineer

About Artyreal

Artyreal is a creative and software studio. We produce design and branding, software products, and video production, across multiple verticals and for clients worldwide. Our team blends storytelling, design, and technology to build work that people actually want to watch and follow. We value creativity, ownership, clear communication, attention to detail, and a strong willingness to learn.

Role overview

We’re looking for an AI Engineer to build the AI products and automations we ship for our own ventures and for clients across Pakistan. You’ll design and build LLM-powered apps, AI agents, RAG systems, and workflow automations, and take them from prototype to production. This is a hands-on, high-ownership role. You won’t spend a year tuning one model in isolation. You’ll ship real systems that real businesses use, across a wide range of problems. It’s ideal for an early-career engineer with strong fundamentals, a few things already shipped, and a genuine obsession with what AI can do right now.

Key responsibilities

• Build LLM-powered applications and AI agents using models from OpenAI, Anthropic, Google, and open-source providers.
• Design and implement RAG pipelines: document ingestion, chunking, embeddings, vector search, and retrieval tuning.
• Build agentic workflows with tool calling, structured outputs, memory, and multi-step reasoning.
• Create workflow automations that connect AI to business tools (CRMs, WhatsApp, email, Google Workspace, Slack) using n8n, APIs, and webhooks.
• Build web apps, internal tools, and MVPs end to end using Claude: Claude Code for AI-assisted development and the Claude API to power AI features.
• Integrate AI features into web and mobile products alongside our product and frontend teams.
• Write and iterate on prompts and system instructions, and version them like code.
• Set up evaluations to measure accuracy, hallucinations, and regressions before anything ships.
• Monitor production systems for cost, latency, and reliability, and optimize where it counts.
• Handle data responsibly: secure API keys, protect client data, and guard against prompt injection.
• Scope technical solutions with the team and explain trade-offs clearly to non-technical clients.
• Stay on top of new models, tools, and techniques, and test what’s worth adopting.

Required qualifications

• 0 to 2 years of experience building software, with at least one AI or LLM project you’ve shipped (work, freelance, hackathon, or personal).
• Strong Python skills. You write clean, tested, production-ready code.
• Hands-on experience with LLM APIs (OpenAI, Anthropic, Gemini, or similar), including tool calling and structured outputs.
• Understanding of RAG, embeddings, and vector databases (pgvector, Pinecone, Qdrant, Chroma, or similar).
• Experience building REST APIs with FastAPI, Flask, or Node.js.
• Comfortable shipping full apps with AI coding tools, especially Claude Code. You use AI to build faster without losing code quality.
• Working knowledge of SQL and at least one database (PostgreSQL, MongoDB, or similar).
• Comfortable with Git, debugging, and working in a shared codebase.
• Clear communication and the ability to explain technical decisions simply.
• Degree in Computer Science, Software Engineering, AI, Data Science, or a related field, or equivalent hands-on experience.

Preferred qualifications

• Experience with agent frameworks (LangGraph, LangChain, LlamaIndex, OpenAI Agents SDK, or Claude Agent SDK).
• Familiarity with Model Context Protocol (MCP) servers and tool integrations.
• Automation experience with n8n, Make, or Zapier.
• TypeScript and Next.js experience for shipping full-stack AI features.
• Docker and cloud deployment (AWS, GCP, Azure, Vercel, or similar).
• Evaluation and observability tools (Langfuse, LangSmith, or similar).
• Running or fine-tuning open-source models (Llama, Qwen, Mistral) with Hugging Face, Ollama, or vLLM.
• Voice AI, speech-to-text, or OCR experience.
• Experience handling Urdu or Roman Urdu text and speech in AI systems.
• WhatsApp Business API or Meta Graph API integrations.
• An active GitHub, open-source contributions, or hackathon wins.

Benefits

• Competitive monthly salary based on skills, plus performance bonuses.
• Ship real AI products used by businesses across Pakistan, not demos.
• Work directly with the founders on product and architecture decisions.
• A wide range of problem spaces across our own products and client work.
• A clear growth path into a senior AI engineer or AI lead role.
• A fast-paced studio environment with real ownership.

Employment type

Full-time, permanent role.

Work arrangement

On-site at our office in Islamabad. Expected commitment: 40 hours/week.