AI Engineer

At Aspect, we’re one of London’s largest property maintenance teams, covering more trades than anyone else and operating 24/7. After more than 15 years of trusted service to thousands of residential and commercial customers, we’re entering an exciting phase of rapid digital transformation. We’re investing heavily in AI and automation to reimagine how we deliver customer service, field operations, and internal workflows.

We’re now looking for a motivated Junior AI Engineer with hands‑on experience in voice AI, retrieval‑augmented generation, and agentic systems to join our growing AI team and help build scalable, production‑grade intelligent systems.

As a Junior AI Engineer, you’ll work alongside our senior AI team to design, develop, and deploy voice‑enabled AI systems and agentic applications that directly impact how we serve thousands of customers and coordinate field engineers across London. You’ll gain deep exposure to production AI systems while contributing meaningfully from day one.

This is a hands‑on engineering role with real ownership: you’ll build, test, and ship AI features into live products used daily across our operations.

• Build and maintain production voice AI agents using platforms such as Retell AI, Live Kit Agents, or similar, ensuring they are scalable, reliable, and optimised for real‑world call volumes.
• Develop and optimise Retrieval‑Augmented Generation (RAG) pipelines to power knowledge‑driven conversations and automate specialist tasks across property maintenance workflows.
• Implement and iterate on multi‑agent architectures (e.g., Lang Graph, CrewAI, Auto Gen) to coordinate complex, multi‑step workflows and automate operational processes.
• Fine‑tune large language models (LLMs) for domain‑specific tasks including call classification, intent detection, and specialised property maintenance knowledge.
• Integrate AI systems with platforms such as Salesforce and internal tools to ensure seamless data flow and operational alignment.
• Design scalable voice agent architectures capable of handling high call volumes with low latency and high accuracy.
• Collaborate with Product, Operations, and Customer Service teams to identify AI use cases and translate requirements into working solutions.
• Monitor, evaluate, and improve deployed AI systems using metrics, logging, and feedback loops to ensure continuous performance improvement.
• Contribute to responsible AI development with attention to fairness, data privacy, and security best practices.

• 1–3 years of experience in AI/ML engineering, with demonstrable work in Generative AI and NLP.
• Strong Python programming skills; familiarity with libraries such as Num Py, Pandas, scikit‑learn, and at least one deep learning framework (PyTorch or Tensor Flow).
• Hands‑on experience building or working with voice AI agents (e.g., Retell AI, Live Kit Agents, Twilio Voice, or similar telephony/voice platforms).
• Practical experience implementing RAG pipelines using frameworks like Lang Chain, Llama Index, or similar.
• Familiarity with multi‑agent frameworks (Lang Graph, CrewAI, Auto Gen) and agentic workflow design patterns.
• Experience fine‑tuning LLMs (OpenAI, LLaMA, Mistral, or similar) for classification, summarisation, or domain adaptation tasks.
• Understanding of vector databases (Pinecone, FAISS, Qdrant) and embedding‑based retrieval.
• Experience deploying AI services on cloud platforms (GCP, AWS, or Azure) with basic understanding of CI/CD pipelines.
• Ability to design voice agent systems with scalability in mind – handling concurrent calls, managing latency, and ensuring high availability.
• Good communication skills and willingness to work collaboratively across technical and non‑technical teams.

• Experience building voice agents at scale (high call volumes, multi‑tenant architectures, or enterprise‑grade deployments).
• Familiarity with Salesforce integration, CRM automation, or customer service platforms.
• Knowledge of prompt engineering best practices, evaluation frameworks, and LLM observability tools (e.g., Lang Smith, Weights & Biases).
• Experience with real‑time streaming, Web Sockets, or telephony…