Machine Learning Engineer (ID: OMvDK4gk)

Machine Learning Engineer, Disinformation AI - Online disinformation is no longer a background problem. It can move markets, damage public trust, distort elections and put organisations on the defensive before they have even understood what is happening.

We are a UK-based AI SaaS start up building software that helps large organisations, governments and NGOs detect, predict and respond to hostile information activity online. The work is serious, technical and genuinely useful. If you want your machine learning skills to sit closer to real-world risk, this role should interest you.

The team now needs a Machine Learning Engineer who has spent at least two years writing serious Python and has built with multimodal, multi-agentic RAG rather than only reading about it. It is a small group, so production code does not sit around for months. Your calls will matter, including work used in sensitive, high-pressure settings.

The work:

A core part of the job is building models that spot, assess and support responses to coordinated disinformation activity. Expect large datasets, Python, model training and validation, plus day-to-day use of PyTorch, Hugging Face, Pandas and NumPy.
You will also help connect ML systems to existing web products and internal services. Some days will be model work. Some will be data quality, pipeline checks, inference performance or debugging behaviour that looks odd in production. We need someone comfortable with the full path from experiment to deployed feature.
You will work closely with data engineers and product colleagues, review code, improve our ML development habits and help us make sensible technical calls as the platform grows.

What we are looking for:

Machine Learning Engineer should have solid Python skills and practical experience with machine learning in a production-minded environment. PyTorch experience will put you in a strong position. If you have also written C++, Rust or similar systems code, even better, though it is not a blocker.

RAG orchestration experience matters here, whether that is LangChain, LangGraph, LlamaIndex or a similar stack. Qdrant, FAISS or another vector database should feel familiar too. Fine-tuning experience with PEFT, LoRA, QLoRA, DPO or ORPO on open-source models would be valuable.

Experience with OpenAI, Anthropic or Gemini APIs, inference tools such as vLLM, llama.cpp or SGLang, and deployment through Docker, Kubernetes and a major cloud platform would all help. We also value knowledge of multi-agent systems, including AutoGen, CrewAI or LangGraph, and multimodal models for image, video or audio, such as Qwen or Whisper.

Why join us as a Machine Learning Engineer:

• Salary from £55,000 per year
• EMI share options, so you own part of what you are building
• Work with a small, fast-moving team where good ideas do not get lost
• Hybrid remote role based in London W10
• Company pension, sick pay, employee stock purchase plan and on-site gym

We are moving quickly on this hire. If the problem feels worth your time, apply now.

Position: Machine Learning Engineer - Job Type: Full-Time - Salary: From £55,000 per year - Location: Hybrid Remote in London