Open Role
LLM Optimization Engineer
Large Language Models are at the heart of what makes Velyon's automations intelligent. As our LLM Optimisation Engineer, you will be responsible for designing, fine-tuning, and optimising the AI models and prompting architectures that power our client-facing agents. You will ensure our AI systems produce accurate, reliable, and contextually appropriate outputs — at scale.
Role Details
What You Will Own
This role is designed for someone who wants practical ownership, direct client impact, and the room to build systems that continue running long after launch.
Design and optimise prompt architectures for production LLM applications (GPT-4o, Claude, Gemini, etc.)
Build and maintain RAG (Retrieval-Augmented Generation) pipelines for client knowledge bases
Evaluate and benchmark model outputs for quality, accuracy, and cost-efficiency
Fine-tune or adapt open-source models where appropriate for specific client use cases
Collaborate with the workflow team to integrate LLM outputs into automation pipelines seamlessly
Monitor LLM performance in production and implement iterative improvements
Research and evaluate new AI models, tools, and techniques as the space evolves
Required Experience
What You Bring
- Deep understanding of LLM architectures and prompt engineering techniques
- Hands-on experience building RAG pipelines (LangChain, LlamaIndex, or equivalent)
- Proficiency in Python
- Experience working with OpenAI, Anthropic, or equivalent APIs in production
- Strong ability to evaluate and improve model outputs systematically
Nice to Have
What Helps You Stand Out
- Experience fine-tuning open-source models (Mistral, LLaMA, etc.)
- Familiarity with vector databases (Pinecone, Weaviate, Chroma)
- Background in NLP or machine learning research
- Exposure to voice AI systems (Vapi, Whisper, ElevenLabs)
Life at Velyon
Ownership, impact, and room to grow
Our culture is shaped by our core values — Growth, Excellence, Innovation and Respect — and reflected in how we collaborate and deliver for our clients worldwide. We are also a certified Great Place to Work, recognised for our commitment to continuous learning, teamwork, and the wellbeing of our people. At Velyon, you won't just have a job — you'll have ownership, impact, and a seat at the table.
Competitive salary aligned with experience and market standards
Flexible remote-first working model
Direct collaboration with the founding team from day one
Fast career growth in a high-impact, early-stage AI company
Hiring Process
A Clear Path from First Call to Offer
Introductory conversation with the founding team
Practical role-specific discussion or portfolio review
Technical or strategic deep dive with the people you will work with
Final alignment on scope, expectations, compensation, and start date
Ready to Build with Velyon?
Send us your profile, portfolio, or a short note about the work you want to do. We care about clarity, ownership, and the ability to build useful things.
Get the latest updates from Velyon.