AI Application Engineer in Bedford

About Veridox

Veridox works with some of the largest insurers, airlines, law firms, and financial institutions in the world to detect fraud in documents and images. When a customer submits a receipt, ID document, or damage photograph, our platform determines whether it is genuine, tampered with, or AI-generated and, critically, explains why with clear, verifiable evidence. We’re replacing manual processes with multi-agent AI that reasons more like a human forensic analyst and we are now scaling rapidly.

Our Culture

We are, first and foremost, builders. Our team is mature and professional, placing significant emphasis on humility, collaboration, and accountability. We operate as a collective: the success of the team takes precedence, we solve challenging problems together, and we share the credit uniformly. We seek individuals driven by the success of the overall system and the team, rather than individual accolades. While our internal environment is relaxed and highly collaborative, we maintain strong professionalism in our routines and customer interactions. We want autonomous, reliable engineers who take immense pride in their craft and aren’t afraid to take ownership directly in front of large enterprise customers.

The Role

You will be the key owner for the development and deployment of our agentic document analysis systems. Your work will extend significantly beyond simple prompt engineering; you will be architecting the resilient infrastructure required for our Large Language Models (LLMs) and Small Language Models (SLMs) to reliably ingest, process, and reason over complex documents at scale. We are looking for an engineer who not only understands how to construct robust production systems today but also possesses the strategic vision to guide our transition toward a fully agentic, skill-utilising platform in the future.

Key Responsibilities

• Build, deploy, and maintain highly resilient, asynchronous workflows using Azure Durable Functions (Python 3.14).
• Develop and iterate on agentic systems utilising LiteLLM, Vertex AI (Gemini Preview & GA Models), and Gemini 2 Embeddings.
• Design and optimise retrieval pipelines using AWS OpenSearch.
• Containerise applications using Docker and manage infrastructure-as-code deployments via Terraform.
• Collaborate closely with our platform teams, competently engaging in discussions about system architecture and integrating with NodeJS APIs.
• Lead technical discussions with enterprise customers, owning your systems and explaining complex AI concepts with clarity and professionalism.

What You’ll Bring

• Deep expertise in Python and cloud-native architecture (specifically handling the complexities and nuances of Azure Durable Functions).
• Proven experience building and deploying production-grade LLM applications, moving substantially beyond experimental setups or Jupyter notebooks.
• Strong understanding of agentic design patterns (including tool calling, routing, and state management).
• A true "builder's mentality": you must be trusted with the autonomy to take a feature from initial concept through to a hardened production deployment.
• Exceptional professional courtesy, communication skills, and a willingness to stand confidently behind your work in front of clients.

Why Join Us?

• Build systems that matter: We aren’t building toy wrappers. We are deploying AI into high-stakes environments where your engineering directly solves complex, enterprise-scale problems.
• Autonomy with support: We hire experts and trust them. You will have the autonomy to design, build, and deploy end-to-end, backed by a collaborative team that has your back when things get difficult.
• The best of both worlds: Enjoy a safe, relaxed, and genuinely fun internal culture combined with a fierce commitment to professional excellence and customer success.
• No red tape, just engineering: We minimise bureaucracy so you can focus on what you do best: writing great code, building resilient cloud architecture, and pushing the boundaries of AI infrastructure.