Mid Full-Stack Engineer (React) — AgentX

Who we are

AgentX is a platform for building and running production AI agents - and a team that ships real systems on top of it for clients who need more than a demo.

The platform is the infrastructure: orchestration, memory, tools, evals, deployment - the layer that makes agentic systems actually work in production. On top of it, we build end-to-end solutions for our clients. One of the systems running on AgentX today is KsięgoBoty - a multi-agent product used daily by hundreds of Polish accounting firms. Paying customers, production load, real consequences when something breaks.

We're small, technical, AI-heavy, and we ship.

Who we're looking for

A solid React engineer who actually understands what's happening under the hood, and uses AI tools daily without outsourcing their brain to them.

You'll join the team that owns the user-facing layer of our products: the dashboards, the operator panels, the customer-facing apps that sit on top of the agents. We need someone we can trust to take a feature, build it well, and ship it without it coming back broken three days later.

This isn't a flashy role and we're not looking for a flashy person. We need craftsmanship.

If you read a ticket and your first instinct is "what's the data model, what's the component structure, where does state actually live" - keep reading.

If your first instinct is "I'll just vibe-code it with Cursor and hope the PR passes" - this isn't for you, and that's fine.

What we mean by "good React engineer"

• You write TypeScript well: types help you, they don't slow you down

• You understand React deeply enough to debug a re-render problem without guessing

• You can structure a non-trivial frontend so it doesn't turn into a tangle in six months

• You've worked with a real backend: REST, auth, data sync, edge cases — and you can hold both ends of a feature in your head

• You read other people's code carefully and review like an adult: kindly, but honestly

• You are comfortable using AI tools daily, but you do not outsource your judgment to them

• You know how to check AI-generated code instead of trusting it because it “looks right”

~3–5 years of commercial experience is roughly where we expect this person to sit. Less is fine if you're sharp; more is fine if you actually want a Mid role.

How we work with AI

We use AI heavily. Claude Code, Codex, Cursor, MCP, agent workflows: they're a normal part of how we work, the same way Git or Docker is. You'll be expected to use them the same way: not as a gimmick, not as a shortcut around understanding the codebase, but as leverage.

That means: break work into clear chunks, guide the tools through implementation, review their output critically, fix what's wrong, and own the final result. You should be faster with these tools than without them - and you should know exactly where they're about to lie to you.

AI doesn't turn a bad engineer into a good one. It just makes them ship faster, which is worse. So:

We do not want people who only know how to prompt.

We also do not want people who proudly refuse to use AI and treat it like autocomplete with better marketing.

The right person sits in the middle: strong fundamentals, strong taste, strong ownership - multiplied by modern AI tooling.

What you'll actually own

• Frontend, day-to-day. React, TypeScript, Tailwind, Next.js. Real product surfaces - operator dashboards, customer panels, internal tools that real people use to do real work.

• Fullstack when needed. You won't live exclusively in the frontend. Node/Python on the backend when a feature needs it - you don't have to be an expert, but you have to be willing.

• Quality. The thing you ship works, doesn't break the next thing, and doesn't make the codebase worse. We notice this.

• Working with the team. Code reviews, architecture decisions - we figure out the harder calls together. You're not coding in a vacuum.

The stack

TypeScript / React / Next.js · Tailwind · Node.js · Python (some) · MongoDB · AWS · Claude / Cursor / MCP for tooling.

We don't use Go.

Nice-to-have (not required)

• You've built a side project that you actually shipped, not just started

• You've worked with LLM APIs in production (not "I made a chatbot in a tutorial")

• You care about UX - not as a designer, but as someone who notices when something feels wrong
  

What we offer

• 13 000 – 18 000 PLN netto + VAT, B2B

• Remote-friendly, Poland-based

• A small team that ships and doesn't waste your time

• Direct work with the founders - no Jira theater, no three-round groomings, no PM filtering your ideas

• Modern AI tooling fully paid for (Claude Max, Cursor, Codex, whatever helps you work better)

• Real ownership of the frontend layer of a product people use every day

How the process works

1. You send us stuff recruitment@agentx.so

• Your CV — yes, we read them

• One thing you built end-to-end you're proud of — and what you'd change about it today

• One AI product (yours or someone else's) you think is genuinely good, and why

• Your take: a client says our agent is "mostly right but wrong in ways that scare them." What do you do first?

2. We reply within 2-3 days — either "let's talk" or "not this time, here's why." No black hole.

3. We meet — 30-45 min call with the founder. We talk about your work, you ask us hard questions. If the chemistry isn't there, no hard feelings.

4. Problem-solving session (top 2-3 candidates) — we give you a real problem we're working on. You come back with how you'd approach it: your thinking, your trade-offs, your questions. We're not watching over your shoulder. We want to see how you think.

5. Decision within 2-3 days. If yes — you start when you're ready. Tomorrow, next week, after notice period. Your call.