LLM R&D Engineer (Remote / Full time)

Location: Remote (main office in Copenhagen, but we have a Ukrainian team)
Seniority Level: Early-career
Job Type: Full-time
Department: Engineering / Infrastructure

Languages: Fluent English and Ukrainian (daily communication and team calls in English)

💬 Strong spoken English is required — you’ll be actively communicating with the international team during standups, technical discussions, and syncs.

🇺🇦 Ukrainian is a plus for informal internal comms with the local team, but English will be your primary working language.

🧠 About Manifold AI

Manifold AI builds intelligent systems powered by large language models. We partner with product teams to create practical tools — from autonomous agents to backend automations and integrated workflows.

We’re a small, experienced team that values autonomy, curiosity, and clarity. We move quickly, prototype often, and focus on solving real problems with the latest in applied AI. If you enjoy building useful systems, thinking independently, and exploring new technical patterns — you’ll feel right at home.

🛠️ About the Role

We’re looking for an Agentic LLM R&D Engineer to help us develop the building blocks behind our AI colleagues — reusable, modular components that support both internal experimentation and customer-facing systems.

You’ll work across short prototyping cycles to turn loosely defined ideas into working code. Some days you’ll explore orchestration patterns and retrieval flows; other days, you’ll be wiring APIs or testing coordination logic between agents.

This is a great fit for someone early in their career who loves LLM Agent systems design, experiments independently, and enjoys the challenge of structuring complex systems from the ground up.

🧩 What You’ll Be Doing

You’ll focus on building infrastructure and logic that powers intelligent behaviors — systems that go beyond basic prompting and into orchestration, memory, and tool integration.

Your responsibilities will include:

• Building internal components for LLM-based workflows and modular AI features
• Prototyping orchestration patterns and agent coordination logic
• Writing clean, maintainable code to integrate tools, services, and model outputs
• Contributing to early product infrastructure, from APIs to vector search
• Translating research, product intent, and open-ended ideas into working systems
• Staying up to date with emerging practices and sharing learnings with the team
• Mostly writing in python and/or typescript

✅ What We’re Looking For

Technical Skills

• 1–3 years of software development experience
• Experience working with LLM APIs or building language model—powered features
• Familiarity with frameworks like LangChain, LangGraph, Botpress, or similar
• Exposure to agentic workflows, multi-step reasoning, or RAG pipelines
• Comfortable using Git, APIs, and command-line tools

Nice to Have

• Personal or open-source projects with LLMs, agents, or automations
• Experience with vector databases or document retrieval workflows
• Participation in hackathons, research groups, or self-driven technical exploration
• Familiarity with coordinating across multiple services or models
• Student club activities
• Experience in data science

Mindset

• Systems thinker — you enjoy building composable, extensible tools
• Curious and independent — you find joy in experimentation and learning
• Pragmatic and product-aware — you want to ship things that work
• Comfortable with ambiguity and open-ended problem spaces
• Strong communicator — able to share ideas, tradeoffs, and technical insights clearly