Machine Learning Engineer

£50,000 + EMI Share Options

London-based | Hybrid (2 days/week in-office)

We live in an era of information warfare. From state-sponsored disinformation campaigns to coordinated attacks on corporate reputation, the digital landscape has never been more hostile.

Valent is a UK-based AI SaaS start up dedicated to detecting, predicting and responding to online Disinformation. We equip large organizations, governments, and NGOs with the tools they need to navigate adversarial environments and defend their integrity.

We are looking for Machine Learning Engineer with a minimum of two years expertise in Python and multimodal multi-agentic RAG.

What You Will Do

• Develop and deploy machine learning models to enhance our disinformation detection.

• Work with large datasets to train and validate models using Python and popular machine learning libraries.

• Collaborate with data engineers to ensure data integrity and efficient data pipelines.

• Integrate machine learning algorithms into existing web applications and services.

• Participate in cross-functional teams to define, design, and ship new machine learning features.

• Conduct code reviews and enhance the machine learning development process.

• Troubleshoot, debug, and upgrade machine learning systems.

• Write clean, efficient, and maintainable code while implementing security and data protection.

Core Language & Frameworks

• Programming Language: Python, C++/Rust (optional)
• ML/DL Frameworks: PyTorch (Preferred), Hugging Face (Transformers, Datasets, Tokenizers)
• Data Processing: Pandas, NumPy

Generative AI & LLMs

• LLM Orchestration (RAG): LangChain, LangGraph, LlamaIndex or similar
• Vector Databases: Qdrant, FAISS, or similar
• Fine-Tuning: Experience with fine-tuning techniques (e.g., PEFT, LoRA, QLoRA) on open-source models (e.g., Llama, Mistral), alignment-tuning (e.g. DPO, ORPO).
• APIs: OpenAI, Anthropic, Gemini, etc.
• Inference: vLLM, llama.cpp, SGLang, etc.

MLOps & Deployment

• Containerization: Docker
• Orchestration: Kubernetes (K8s) (for scalable inference)
• Cloud Platform: AWS, GCP, or Azure (experience with S3, EC2/GCE, and a managed Kubernetes service like EKS/GKE is a strong plus)
• Experiment Tracking: Weights & Biases (W&B) or MLflow
• Version Control: Git / GitHub

Specialised (Nice-to-Have)

• Multi-Agent Systems: AutoGen, CrewAI, LangGraph
• Multimodal: Experience with models or techniques for handling images/video (e.g., Qwen3, Whisper, etc.)

Why Join Valent?

£50k base + EMI share options (own a piece of the future you are building).

Impact: You are building software that will make a difference in the world

Small dynamic team with an inclusive work environment

We are hiring fast. Apply now.