Backend Software Engineer – Applied ML & LLM Systems

About the role

Dwelly is a fast‑growing AI‑enabled lettings and property‑management platform looking for a Backend Software Engineer with strong applied machine‑learning experience. You will design and ship production‑grade services that turn unstructured emails, documents and other data into structured, actionable information for real‑estate agencies.

Key responsibilities

• Build backend systems that extract, enrich and summarise data from emails, documents and other unstructured sources.
• Enrich migrated client, landlord, tenant and property records with insights derived from communication history.
• Develop solutions that summarise a client’s full email history and surface relevant context within the Dwelly platform.
• Implement reliable NLP/ML‑backed services that operate on messy real‑world data.
• Improve retrieval and ranking pipelines using RAG, BM25, embeddings, hybrid search and reranking techniques.
• Define quality metrics, create evaluation datasets and establish feedback loops for extraction, summarisation and retrieval.
• Build Python APIs and services with frameworks such as FastAPI, Django or Flask, ensuring clear error handling, observability and maintainability.
• Collaborate closely with engineering, product and operations teams to translate business problems into scalable automation.

Required profile

• Strong Python backend engineering experience.
• Hands‑on experience with API frameworks (FastAPI, Django, Flask or similar).
• Production experience with NLP, machine‑learning, information extraction, retrieval, ranking or summarisation systems.
• Proven ability to take research prototypes to production‑ready services.
• Comfort working with messy, real‑world data and iterating based on evaluation results.

Required skills

• Python
• FastAPI / Django / Flask
• Natural Language Processing (NLP)
• Machine Learning (ML) and Large Language Models (LLM)
• Information extraction and summarisation
• Retrieval techniques (RAG, BM25, embeddings, hybrid search, reranking)
• API design and development
• Observability and error‑handling in production systems