Logo for HYRED

Agent Engineer

Roles & Responsibilities

  • Build LLM-powered agents from scratch, including planning loop, tool usage, memory management, and execution
  • Proficiency in JavaScript/TypeScript (or another language) with full-stack capability across the agent stack
  • Experience deploying apps with modern cloud tech (Vercel, AWS, Docker + VPS) and integrating vector memory stores (Pinecone, Weaviate) and graph databases
  • Strong product mindset, curiosity, adaptability, and ability to thrive in a fast-paced startup with AI-native tooling

Requirements:

  • Build and ship end-to-end LLM-powered agents, connecting planning, tool usage, execution, and memory management
  • Design and implement the shared memory layer (graph DB, vector store, key-value store) to enable real-time knowledge sharing across agents
  • Develop, debug, and deploy across the full agent stack, from orchestration to state management and tool execution
  • Collaborate with cross-functional teams to deliver weekly features and improve user outcomes in a high-velocity startup culture

Job description

About the client

Our client is a self-improving memory layer for your AI coding agents—create, retrieve, manage vibe-coding best practices across projects and teams. 

Big Picture

Recent breakthroughs in automatic programming, often called "vibe coding," have enabled AI agents to handle increasingly sophisticated software development tasks. Systems like Cursor, Windsurf, and Claude Code showcase these advanced capabilities. 

Yet, a significant challenge remains: most developer agents operate independently, unable to share their reasoning, problem-solving approaches, or interactions with humans. As a result, agents frequently repeat similar efforts, essentially reinventing the wheel each day.

They solve this problem by providing a shared memory layer for developer agents. It enables AI agents to instantly access and utilise the experiences, context, and solutions previously discovered by other agents. 

By doing so, they empower agents to collaboratively overcome complex tasks in real-time, eliminating redundant reasoning loops and accelerating productivity.

At its core, the company leverages a layered datastore architecture that integrates graph databases to capture relationships between entities, vector stores for rich content representation, and key-value stores for efficient context retrieval. This unified design ensures agents quickly access relevant knowledge and past experiences, enabling smarter decision-making and more cost-effective problem-solving.

Here's what they would love to see in you:

  • You know how to build LLM-powered agents from scratch.
  • You understand the core concepts of agents, tool usage, memory management, and how to wire everything together — from the planning loop to execution.
  • Strong product mindset.
  • You think beyond just code — you care about the why behind what they are building, and want to make users happy.
  • Fluent in one programming language - Could be Python, JavaScript, Go, Rust, you name it. They are currently working with JS/TypeScript, but believe any good dev can pick up new languages when needed.
  • Full-stack capable across the agent stack.
  • You get the full picture — from LLM orchestration, vector memory (like Pinecone or Weaviate), tool/action execution (MCPs), to managing agent state. You can build, debug, and ship across the whole flow.
  • You ship.
  • You're familiar with deploying apps using modern cloud tech. Whether its Vercel, AWS, or Docker + VPS — you've got it.
  • High-velocity mindset.
  • They ship weekly and like small iterations, fast feedback, and improving constantly.
  • Curious & adaptable.
  • You're not afraid to dive into unknown territory and learn fast.
  • AI-native developer.
  • You already use tools like Cursor, Windsurf, Cline, or Claude to speed up your workflow. (they do too!)
  • They dont care about years of experience — they care about what you can do, how you think, and whether you're ready to build something awesome with them.

What you can expect joining them

  • Build something global, with a solid team.
  • You'll be working alongside experienced engineers and product folks who have shipped real things before — and are now aiming even bigger.
  • True startup energy.
  • Fast-moving, scrappy, and high-trust. They are VC-backed, but still early enough that you'll make a real dent.
  • Competitive compensation — and equity.
  • We believe in rewarding the people who help build the future.
  • No politics, just real collaboration.
  • They keep things open, honest, and direct. No fluff, no games — just great people building together.
  • Grow like crazy
  • They care about speed — not just in shipping code, but in helping you grow technically, professionally, and personally.

Related jobs

Other jobs at HYRED

We help you get seen. Not ignored.

We help you get seen faster — by the right people.

🚀

Auto-Apply

We apply for you — automatically and instantly.

Save time, skip forms, and stay on top of every opportunity. Because you can't get seen if you're not in the race.

AI Match Feedback

Know your real match before you apply.

Get a detailed AI assessment of your profile against each job posting. Because getting seen starts with passing the filters.

Upgrade to Premium. Apply smarter and get noticed.

Upgrade to Premium

Join thousands of professionals who got noticed and hired faster.