AI Engineer, Forensic & Financial Crime Technology Advisory and Data Analytics

Our teams in Forensic and Financial Crime provide proactive advice, transformation and operational support to help clients protect their brand and navigate financial crime risks, including money laundering, fraud and corruption.

• Proven experience designing and shipping software or AI solutions end to end, comfortable owning both the architecture and the code;
• Strong system-design skills, able to decompose a problem into front end, back end, APIs, data, caching, memory, async/eventing and logging, and to communicate the design with C4/sequence diagrams;
• Hands-on experience building LLM/GenAI applications: RAG, tool use, orchestration and (ideally) multi-agent patterns, and a clear view on when NOT to use them;
• Practical prompt engineering for production, using structured outputs (JSON schema) and tool/function calling for reliable behaviour;
• Experience evaluating non-deterministic systems: golden datasets, regression testing and eval frameworks, and using tracing/observability to debug them;
• Strong coding ability in Python and/or TypeScript, working within Git/PR workflows with automated testing;
• Data engineering fundamentals: relational, NoSQL and vector stores, and document/ETL pipelines (parsing, chunking, embedding);
• Fluent, responsible use of AI coding assistants while still testing, debugging and owning the output;
• Ability to lead a workstream and mentor others, and to communicate complex designs to both technical and non-technical (including senior client) audiences., * Experience redesigning a business process (ideally in financial crime or a regulated domain) around GenAI rather than automating it like-for-like;
• Familiarity with orchestration and eval tooling such as LangChain/LangGraph, LlamaIndex, Ragas, DeepEval, LangSmith, Langfuse and Azure AI Foundry;
• Experience with vector search (pgvector, FAISS, Pinecone), Redis and message queues;
• Document-intelligence experience (Unstructured, PyMuPDF/pdfplumber, Azure Document Intelligence, AWS Textract) and BI/dashboarding (Power BI, Qlik Sense);
• Experience of working on projects for external clients; and
• Experience of working in or on projects in regulated industries, particularly financial services.

Turning messy, real-world financial crime processes into reliable, production-grade GenAI systems, it’s just another day at the office for our Financial Crime Technology Advisory and Data Analytics experts.

This is your opportunity to join Deloitte’s Financial Crime Technology Advisory and Data Analytics team. Our rapidly growing practice is expanding to meet large and consistent client demand for AI-enabled solutions. The team comprises a group of specialists drawn from a wide range of different professional and technical backgrounds. We apply our combined skills and experience to find new ways of solving our clients’ most complex technology, data, and analysis challenges.

Working with the worlds’ largest and most complex financial organisations, you will own the design and delivery of AI solutions end to end, from system architecture and where the AI actually fits, through agentic workflow design, prompting and evaluation, to building the code and data pipelines that make it work in production. Collaborating with colleagues and industry leaders, you will have an opportunity to grow your industry presence and be recognised as a leader in your domain.

No two projects are the same, however the majority will involve rethinking a financial crime process around GenAI rather than copying it step-for-step, and then engineering a solution that is accurate, measurable and safe to run. Most of the projects will be in the regulated financial services sector.

Away from project work, we offer both formal and informal support for continuous professional development, to accelerate your career progression within a rapidly growing team.

We encourage consideration of flexible ways of working, both formal and informal arrangements that allow for the best outcomes for our people and our clients. If this opportunity is of interest to you with some flexibility, please do discuss it with us.

Connect to your career at Deloitte

Deloitte drives progress. Using our vast range of expertise, we help our clients become leaders wherever they choose to compete. To do this, we invest in outstanding people. We build teams of future thinkers, with diverse talents and backgrounds, and empower them all to reach for and achieve more.

What brings us all together at Deloitte? It’s how we approach the thousands of decisions we make every day. How we behave, our beliefs and our attitudes. In other words: our values. Whatever we do, wherever we are in the world, we lead the way, serve with integrity, take care of each other, foster inclusion, and collaborate for measurable impact. These five shared values lead every decision we make and action we take, guiding us to deliver impact how and where it matters most.

Connect to your opportunity

Using cutting-edge technology, you’ll work alongside colleagues, often in multi-disciplinary teams, helping to solve our clients’ most significant challenges. This is a hands-on role that spans three connected disciplines: architecting the whole solution and deciding where AI fits; designing and evaluating the agentic workflows and prompts; and building the code and data pipelines that bring it to life. You will be expected to move fluidly between the whiteboard and the codebase.

Solution architecture and system design

You will design the whole solution and decide where the AI sits. Specifically, you will:

• Define the modules and how they fit together: front end, back end, APIs and integration layers, authentication/authorisation, data persistence, caching, session and conversation memory, async/event handling, and logging/observability;
• Decide where and how AI is applied: single LLM call vs multi-agent, tool use and orchestration, RAG vs long-context, and the guardrails around each, and justify the trade-offs;
• Produce clear architecture artefacts (C4 model, sequence and component diagrams) and make deliberate choices on data stores (relational vs NoSQL vs vector), caching, and messaging/queues.
• Tools/tech: C4 model, Mermaid, draw.io; REST/OpenAPI, GraphQL; Redis, relational and NoSQL stores, message queues; Docker/containers; orchestration frameworks (LangChain/LangGraph); vector search and retrieval (LlamaIndex, pgvector, FAISS, Pinecone).

AI workflow design, prompting and evaluation

You will turn business processes into reliable agentic workflows and prove they work. Specifically, you will:

• Redesign financial crime processes around GenAI, rethinking the workflow rather than replicating the existing steps one-to-one;
• Write reliable, maintainable prompts inside production systems, using structured outputs and tool/function calling to make behaviour predictable;
• Measure accuracy and quality in non-deterministic systems, building golden datasets, regression suites and automated evals, and tracing behaviour to diagnose failures.
• Tools/tech: process mapping (BPMN 2.0), flow and sequence diagrams, pseudocode; structured outputs (JSON schema, tool/function calling); golden datasets and regression testing; evaluation frameworks (Ragas, DeepEval, Azure AI Foundry evaluations); advanced prompting (chain-of-thought, LLM-as-judge and similar); tracing (LangSmith, Langfuse).

AI-assisted development and data engineering

You will build the solution and engineer the data that feeds it, using AI coding tools while still genuinely owning the code. Specifically, you will:

• Build using AI coding assistants while understanding, testing, debugging and taking full ownership of the code you ship;
• Engineer the databases, document and data pipelines that feed the solution, including ingestion, parsing, chunking and embedding;
• Work within professional engineering practices: version control, code review, automated testing and CI.
• Tools/tech: AI coding tools (GitHub Copilot, Claude Code, Cursor); Python and TypeScript; LLM APIs (tool use, MCP); databases (MSSQL, Postgres, Cosmos DB); Git/PR workflows (GitHub, Azure DevOps); document extraction (Unstructured, PyMuPDF/pdfplumber, Azure Document Intelligence, AWS Textract); chunking and embedding pipelines., Regulation and controls are standard practice in our industry and Deloitte is no exception. These controls provide important legal protection for both you and the firm. We are subject to a number of audit regulations, one of which requires that certain colleagues abide by specific personal independence constraints (e.g., in relation to any financial interests and employment relationships). This can mean that you and your “Immediate Family Members” are not permitted to hold certain financial interests (shares, funds, bonds etc.) with audit clients of the firm, and also prohibitions on certain employment relationships (e.g., you are not permitted to hold a secondary employment role with SEC audit clients of the firm whilst being employed by the firm). The recruitment team will provide further details as you progress through the recruitment process, or you can contact the Independence team upon request.

Connect with your colleagues

“The scale and range of projects you get to work on, because of Deloitte’s capability across all industries and clients, is why I would recommend a career here.” - Jeremy, Risk & Transactions Advisory

Our hybrid working policy

You’ll be based in London with hybrid working.

At Deloitte we understand the importance of balancing your career alongside your home life. That’s why we’ll support you to work flexibly through our hybrid working policy. Depending on the requirements of your role, you’ll have the opportunity to work in your local office, virtual collaboration spaces, client sites and remotely. You’ll get the chance to meet face to face when needed, while you collaborate and learn from colleagues, share your experiences, and build the relationships that will fuel your career and prioritise your wellbeing. Please check with your recruiter for the specific working requirements that may apply for your role.

Our commitment to you

Making an impact is more than just what we do: it’s why we’re here. So we work hard to create an environment where you can experience a purpose you believe in, the freedom to be you, and the capacity to go further than ever before.

We want you. The true you. Your own strengths, perspective and personality. So we’re nurturing a culture where everyone belongs, feels supported and heard, and is empowered to make a valuable, personal contribution. You can be sure we’ll take your wellbeing seriously, too. Because it’s only when you’re comfortable and at your best that you can make the kind of impact you, and we, live for.

Your expertise is our capability, so we’ll make sure it never stops growing. Whether it’s from the complex work you do, or the people you collaborate with, you’ll learn every day. Through world-class development, you’ll gain invaluable technical and personal skills. Whatever your level, you’ll learn how to lead.