Forward Deployed Data Solution Engineer

About this role

About the Role This is not a traditional enterprise architecture role. Accelerant is operating in a period of rapid change. Significant internal systems across finance, operations, and data are being implemented. Business users across the organisation are building AI-enabled capabilities at a pace that exceeds the capacity of central engineering teams to review and govern them. The ways in which software is designed and built are themselves changing, with AI-assisted development and agentic tooling becoming part of everyday practice. The Forward-Deployed Solution Engineer exists to respond to all of this in real time, from inside the teams doing the work, not from a distance. This person is embedded with delivery programmes and product teams. They review decisions and code as they are made, not after the fact. They drive solutions quickly, make independent technical judgements, and hold the quality bar without becoming a bottleneck. They are the link between what the business needs to build and what can actually be operated and maintained at enterprise scale. What You Will Do Be present in the work You will sit inside delivery teams for the duration of engagements, not alongside them. You will attend design and development sessions, review technical decisions as they are being made, and intervene when something is heading in the wrong direction. You will not wait for a review gate. You will be the engineering rigour in the room. Validate and guide continuously You will continuously assess whether what is being built conforms to Accelerant's architecture principles, data standards, and integration patterns. Where it does not, you will propose a better path and help the team take it. Where documented patterns do not yet exist, you will create them. You will not rely on others to produce the standard after the fact. Take decisions quickly You will be capable of making a sound technical decision on the day, with the information available, and moving forward. You will not wait for consensus where it is not needed. You will know when to escalate and when to act. The teams you work with will rely on your judgement. Bridge business-built capability to production Across Accelerant, business users are building capabilities using AI tools, low-code platforms, and scripting environments. Many of these capabilities are genuinely useful and represent Accelerant's direction of travel. Your job is to assess what has been built, understand what it actually does, identify the gaps between its current state and enterprise readiness, and create a clear path to production. You will ensure these solutions have appropriate data contracts, operational ownership, documentation, and supportability before they are relied upon by the business. Own integration design You will design and validate integration patterns between the enterprise systems Accelerant operates, including ERP, CRM, HR, and reinsurance platforms, and the data platform and applications that consume data from them. You will ensure that integrations are clean, well-documented, API-first where appropriate, and built to a standard that does not require the original designer to be present in order to be supported. Drive data product quality and governance You will work with delivery teams and the Data Office to ensure that data products are described at source, that ownership is clearly assigned, and that the canonical data models Accelerant depends on are accurate and accessible. You will understand the difference between a physical implementation and a logical data model, and you will be able to articulate why that distinction matters to the people building on top of these products. Make knowledge durable You will ensure that technical decisions are recorded, that architectural decision records exist and are accessible, and that the knowledge required to support what Accelerant builds does not live only in individuals. When you leave an engagement, the next person should be able to understand what was decided, why, and what the constraints were. What You Will Not Do You will not write lengthy architecture documents that no one reads. You will not operate as a governance gate that slows delivery without adding value. You will not wait for others to bring problems to you. You will not review things in isolation from the teams building them. You will not produce opinions without evidence. Skills and Experience Technical foundation You have strong practical knowledge of cloud platforms, particularly Azure, AWS, and understand how enterprise applications are integrated, operated, and maintained at scale. You are comfortable with API design and integration patterns, including REST, event-driven architectures, and MCP-style tool connectivity. You understand data product concepts, lineage, ownership, and what it means to describe data at source. You can read and review code written by others and provide substantive feedback on it. AI and modern development practice You understand how AI-assisted and agentic development approaches are changing how software is built. You have practical experience with at least one AI development framework and understand the governance considerations that apply when AI-generated or AI-assisted capabilities are moved into production. You have a view on what makes an AI-enabled capability enterprise-ready and can assess a prototype against that standard. Domain awareness You have experience working with enterprise SaaS platforms, particularly in the finance, operations, or insurance domains. You understand how systems like ERP and CRM platforms model their data, why direct integration to application databases creates problems, and what a canonical data model is trying to solve. Experience with D365, Salesforce, or similar platforms is relevant. Communication and delivery You can communicate clearly with both engineers and business stakeholders, and you adjust your register accordingly. You can explain a technical constraint to a