Kỹ sư AI at DZMM

We are a team developing AI-powered content products, with our main platform being web-based. Our offerings include AI chat, image generation, UGC mini-games, novels, and forum modules. We have real users, genuine traffic, and actual revenue—this is not an early-stage project pitching with PowerPoints, nor are we an outsourcing service. Our product line grows monthly, and with endless tasks, we're looking to expand our team. Our team is small and flat, with no hierarchical reporting or weekly meeting culture. What you build could go live the next day and be used by tens of thousands of users. This role is unique—here’s how it works: Our development process involves humans writing task tickets and AI agents writing code. You won’t be manually coding business logic daily. Instead, you’ll investigate user feedback or requirements, distill them into clear task tickets, assign them to AI agents, oversee execution, review PRs, test, and monitor post-launch data. The agent writes the code, but you’re responsible for its correctness. Your daily tasks will include:

• Investigating user feedback or requirements by reading code, checking logs, querying databases, and consulting stakeholders—this often takes the most time, and that’s intentional.
• Writing concise task tickets specifying what needs to be done, deliverables, and constraints.
• Assigning tasks to AI agents, monitoring progress, and stepping in to correct deviations or restart if necessary.
• Reviewing PRs—agent completion doesn’t mean it’s done; green tests don’t guarantee correctness. You’ll scrutinize diffs and test manually.
• Deploying and verifying whether the issue is truly resolved.

Tech Stack: Our web stack includes TanStack Start, React, TypeScript, TanStack Router, and Query for routing and data fetching. APIs use tRPC with zod validation, databases are Postgres on Supabase (accessed via Drizzle), authentication via Supabase Auth, Redis for caching, jotai for client state, NestJS for standalone services, and Docker with CI/CD for deployment. You don’t need to know the entire stack upfront—we value your ability to learn and judge agent output critically. Note: We’re also hiring Senior Engineers (salary: 4,000–6,000 USD/month) with 3+ years of hands-on experience, monorepo proficiency, and agent workflow expertise. If you qualify, mention "Senior Role" in your application.

Text chat → mock interview (no algorithms). We’ll simulate a real issue—you’ll interact with us as if we’re your AI agent terminal, demonstrating how you’d investigate, ticket, and verify. Takes 1–2 hours; successful candidates discuss salary afterward.