AI / ML / LLM Engineer | Python | PyTorch | Hugging Face | Agentic AI

Description Our client is an established technology company developing advanced voice and data-processing solutions for telecommunications providers and government organisations. Its platforms process large volumes of IP, voice and communications data, supporting complex analytics and operational use cases across international markets. The company is now looking for an experienced AI/ML Engineer to help develop, fine-tune and deploy advanced machine-learning capabilities across natural language processing, large language models and related data modalities. 𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝗶𝗲𝘀 • Design, fine-tune and evaluate Large Language Models for applications such as text classification, entity extraction, summarisation, sentiment analysis and conversational AI. • Build and improve domain-specific training datasets, including data preparation, synthetic-data generation, annotation quality and validation. • Develop reliable machine-learning pipelines for text processing, model training, evaluation and production inference. • Apply parameter-efficient fine-tuning techniques such as LoRA, QLoRA, adapters, prompt tuning and instruction tuning. • Explore Agentic AI patterns and integrate models with external tools, data sources and enterprise workflows. • Develop robust model-evaluation frameworks using automated benchmarks, qualitative review and human-in-the-loop assessment. • Improve model performance through quantisation, pruning, distillation, inference optimisation and efficient deployment. • Collaborate with engineering, product and technical teams to integrate AI capabilities into secure production environments. • Evaluate emerging foundation models and recommend suitable architectures for specific product and customer requirements. • Contribute to reproducible ML development practices, experiment tracking, technical documentation and production monitoring. 𝗧𝗵𝗲 𝗢𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝘆 This role offers the opportunity to work on technically challenging AI applications involving large-scale communications data, language processing and advanced model deployment. The successful candidate will contribute to AI capabilities used in complex telecommunications and government environments while working with an international team on emerging applied-AI use cases. Occasional travel may be required for customer meetings, conferences or collaboration with other offices. Requirements Bachelor's or Master's degree in Computer Science, Artificial Intelligence, Machine Learning, Data Science, Mathematics or a related discipline. Four or more years of experience in machine learning engineering, applied AI or deep-learning development. Strong proficiency in Python, PyTorch and the Hugging Face ecosystem. Practical experience working with transformer architectures and Large Language Models. Hands-on experience with model fine-tuning, evaluation and performance optimisation. Experience with techniques such as LoRA, QLoRA, PEFT, instruction tuning or comparable adaptation methods. Good understanding of tokenisation, embeddings, attention mechanisms, context management and LLM inference. Experience deploying machine-learning models into production environments. Familiarity with Docker, model tracking, version control and reproducible ML workflows. Strong analytical and problem-solving skills, with the ability to evaluate model quality and improve performance. Excellent written and spoken English. Desirable Agentic AI and tool-enabled workflows Model Context Protocol integration Automatic Speech Recognition and speech-to-text Whisper, Wav2Vec2 or related speech models Multimodal models combining text, audio or visual information Diffusion models and synthetic-data generation ONNX, TensorRT, quantisation or inference optimisation Distributed or multi-GPU model training Kubernetes, MLflow, Ray or similar MLOps technologies Real-time inference and streaming-data environments Privacy-preserving, explainable or security-sensitive AI Telecommunications, voice analytics or communications-data applications