ABOUT US

Via an integrated set of capabilities, Transform, Tech & Talent, we deliver pioneering solutions, helping our clients navigate complexity, accelerate business change, and establish market advantage via the effective adoption of technology.

Abstract headquartered in the UK with proven fulfilment capabilities across Europe, APAC & GCC. Our global team is made up of passionate, highly skilled individuals with experience spanning software development through to technology strategy development.

We work alongside leadership teams, prioritising innovation and collaboration to enhance businesses and drive growth.

THE ROLE

As an AI Engineer, you’ll design, build, and deploy AI features that solve complex problems and deliver real value to end users. Working across a variety of client projects, you’ll apply your expertise in areas such as conversational AI, image and data classification, and scalable model deployment, bringing cutting-edge AI to production environments at scale.

You’ll collaborate closely with product managers, domain experts, and engineering teams to ensure the AI systems you build are robust, safe, and intuitive. This is a hands-on role where you’ll be writing code, building pipelines, and iterating quickly based on feedback from real users and stakeholders.

RESPONSIBILTIES

• Lead the development of AI solutions across multiple client engagements.
• Collaborate with clients to understand business challenges and translate them into AI-driven solutions.
• Work with client data teams to ingest, clean, and prepare data for modeling.
• Build and integrate AI solutions using LangChain and similar orchestration frameworks for conversational AI.
• Design and optimise Retrieval-Augmented Generation (RAG) pipelines, including embedding generation, vector database configuration, and retrieval tuning to improve accuracy, latency, and relevance in conversational AI.
• Prompt optimisation of closed source LLMs, or post-training of open-source LLMs to build new or improve existing agentic workflows
• Design and run evaluation frameworks to measure model safety, accuracy, and performance, including recall, precision, F1-score, and other relevant metrics for both conversational AI and classification models.
• Set up and maintain MLflow for experiment tracking, model versioning, and performance monitoring across AI projects.
• Own and improve MLOps pipelines to speed up experimentation, deployment, and monitoring.
• Ship AI models into live production environments and iterate rapidly based on usage, feedback, and metric-driven insights.
• Write clean, testable, and scalable code in Python and integrate with backend services.
• Stay up-to-date with the latest AI research and tools, applying them pragmatically to solve real problems.
• Design and refine prompts, workflows, and evaluation frameworks for autonomous AI financial analysts.
• Conduct rigorous, finance-specific performance tests to ensure AI output is accurate, reliable, and decision-ready.
• Collaborate with the engineering team to integrate AI agents into production systems using Node.js, TypeScript, and Python.
• Identify and implement improvements that increase model reliability and reduce error rates in financial contexts.
• Execute projects from concept to deployment, ensuring deliverables meet both technical and financial industry standards.
• Present findings, model performance, and recommendations in client meetings, communicating technical concepts to non-technical stakeholders.
• Support client teams with onboarding, training, and documentation.
• Incorporate client feedback into model refinement and workflow improvements.

REQUIRED SKILLS AND EXPERIENCE

• 3–5 years of software engineering or AI-related experience, with experience shipping production models.
• Ability to write clean, testable, and scalable code with Python,C# or JS.
• Ability to adapt to different industry domains.
• Experience with LangChain/LangGraph or similar AI orchestration frameworks.
• Proven experience in designing, implementing, and optimising RAG-based systems, including vector databases (e.g., pgvector, Weaviate, Qdrant, Pinecone).
• Experience of image model training, deployment and inference using frameworks such as Pytorch/Tensorflow and cloud-based tools such Azure AI Foundry and OneReach.ai.
• Proficient in MLflow (or similar tools) for tracking experiments, managing models, and maintaining reproducibility.

• Good understanding of conversational AI evaluation , with the ability to run and interpret tests that improve recall, precision, and F1-score, and balance these metrics for optimal customer experience.
• Knowledge of LLM post-training using SFT and RL using frameworks such as TRL.
• Solid understanding of MLOps (CI/CD for models, data pipelines, monitoring, retraining).
• Experience running evaluation and validation processes for AI models in production.
• Comfortable with GitHub workflows, Docker, and cloud deployments.
• Strong communicator, able to explain AI concepts to both technical and non-technical audiences.
• Proficiency in Node.js, TypeScript, and Python.
• Prior experience evaluating AI agents or a strong understanding of their functionality.
• Strong problem-solving skills, with the ability to work on multiple projects.
• Experience with multi-modal AI (text, images, video) is desirable
• Experience operating healthcare, veterinary, or other high-trust domains would be beneficial.
• Reasonable understanding of AI safety and ethics in production systems would be ideal.
• Azure AI Engineer Certification/Azure Developer or Architect certification is a plus.

REMUNERATION AND BENEFITS

• Base salary commensurate with experience.
• Vibrant Leeds City Centre office location.
• Fully stocked communal kitchen with access to free food and drink.
• Games area, complete with pool table.
• 25 days holiday rising an extra day a year up to 30 days.
• Flexible/Hybrid working.
• Employee Assistance Programme.
• Benefits platform offering a range of retail and entertainment discounts.