Planso builds AI-powered decision-intelligence products for the global architecture, engineering and construction industry. We work on complex problems involving enterprise data, documents and domain-specific reasoning.
Responsibilities
Build and evaluate LLM-powered product capabilities.
Develop pipelines for document understanding, information extraction and retrieval.
Experiment with embeddings, reranking, RAG and agentic workflows.
Create evaluation datasets and measure accuracy, relevance and reliability.
Analyse failure cases and improve prompts, models and guardrails.
Work with engineers and domain specialists to take prototypes into production.
Maintain clear documentation of experiments and technical decisions.
What we are looking for
Strong Python and programming fundamentals.
Understanding of machine learning, NLP or information retrieval.
Ability to work with structured and unstructured data.
Understanding of evaluation metrics such as precision and recall.
Strong problem-solving ability, ownership and curiosity.
Good to have
Experience with PyTorch, Hugging Face, LangChain, LangGraph or LlamaIndex.
Exposure to vector databases, knowledge graphs or document-processing systems.
Experience building APIs using FastAPI.
Relevant internship, academic or personal projects. What you will gain
Experience building production-grade applied AI systems.
Exposure to enterprise AI and complex real-world data.
Opportunity to work closely with the founding and engineering teams.
Significant ownership and a fast learning environment.
Application
Share your résumé, GitHub or portfolio, and a short note about an AI project you built—including one challenge you encountered and how you addressed it.