AI Enablement Engineer

Gulliver's Games; Şişli; On-site; TRY 300,000 - 500,000; Full time

Benefits offered by this job

Mentorship from founder

Top-tier AI tooling subscriptions

On-site office Esentepe

Job summary

Gulliver’s Games in Istanbul seeks a hands-on engineer to own AI at the company: the tools, skills, workflows and evals that let every person in the company get real work done with AI agents; reliably, safely, and without reinventing setups.

You will work directly with the founder, who is hands-on with this stack today, and hand the day-to-day ownership of it over to you. This is a role for builders who ship.

Qualifications

• No minimum years of professional experience required.
• Solid technical foundation: CS/engineering degree or equivalent self-taught depth.
• Hands-on with coding agents and AI tooling.
• Builder’s mindset: break down problems, automate, ship.
• Clear communicator with English documentation and prompts.

Responsibilities

• Build and maintain the internal AI toolkit: agents, skills, MCP servers and integrations.
• Create eval suites, tracing and fix failure patterns in agent runs.
• Run the infrastructure behind it: laptops, Mac minis and VMs with safe access.
• Enable the team to use AI effectively from day one.

Tools

Claude Code

Codex

Cursor

MCP servers

Job description

Gulliver’s Games is an Istanbul-based mobile game studio (Word Tiles Go and other titles). We run a small team that uses AI agents across everything: game development, backend and live-ops, data, marketing and internal tools. Our internal AI stack which is Claude Code, Codex, agents, skills, MCP servers, multi-agent workflows, and automation running on local machines and cloud VMs — has grown to the point where it needs a dedicated owner.

The Role

This is ahands-on engineering role, not an advisory or strategy function. You will own the work for AI at the company: the tools, skills, workflows and evals that let every person in the company get real work done with AI agents; reliably, safely, and without reinventing setups.

You will work directly with the founder, who is hands-on with this stack today, and hand the day-to-day ownership of it over to you.

What You’ll Do

• Build and maintain the internal AI toolkit:our agents, skills, MCP servers and integrations with our systems.
• Evals and reliability:eval suites, tracing and fixing failure patterns in agent runs.
• Run the infrastructure behind it:the laptops, Mac minis and VMs our agents run on, with safe access.
• Enable the team:help everyone get real work done with AI, from day one.

What We’re Looking For

• No minimum years of professional experience required.We care about how you think and what you’ve built, not your job title history. New graduates, career-changers and self-taught builders are very welcome.
• Solid technical foundation:A degree in CS, engineering, math or a related field, or equivalent self-taught depth.
• Hands-on with coding agents.You’ve used Claude Code, Codex, Cursor or similar at work, at school or on weekends.
• Builder's mindset.You break down messy problems, debug patiently and find the simple fix. When you notice friction, you automate it and ship.
• Clear communicator:You explain things simply and have patience for teammates using AI for the first time.
• Working English:you read technical docs, write clear prompts and document your work in English every day.

Hiring Process

• Intro call & Project talk: walk us through something you built — personal, hobby, school or work — and how you solved the hard parts. (30 min).
• Practical case (~3 hours): improve one of our real skills and write an eval that proves it got better. We aim to decide within 2 weeks and we reply to every applicant.
• Meet the team and founder.

What We Offer

• Mentorship from a founder who builds with this stack daily, and room to grow fast.
• Top-tier AI tooling subscriptions and hardware.
• Full-time on-site work from our office location, Esentepe.
• Real ownership: you shape how an AI-native studio works.

How to Stand Out

A link beats a CV. Personal and hobby projects count as much as professional work. The applications we remember show us something real: an agent setup you tuned, a skill, plugin or MCP server, an eval you wrote, a bot, a home-lab automation, or a write-up about something you figured out.