Large Language Models
Relay is model-agnostic, selecting the best model for each task. The better the models get, the more ambitious the experiences we can build.
Audit automation, built by auditors
Relay does the keying, the tying out, the rebuild every time the figures change — so your time goes to the numbers that don't look right.
Latest from the blogCloser to the EvidenceAn auditor's real work is the ten seconds it takes to sign. AI can build the workbook — but only if every figure stays traceable, so scepticism has evidence to land on, not just a conclusion to trust.Under the hood
Relay is built across seven layers, from the AI models up to the audit interface, so the work it produces is accurate, reviewable, and safe to use on a real engagement.
Relay is model-agnostic, selecting the best model for each task. The better the models get, the more ambitious the experiences we can build.
The agentic harness is the orchestration engine turning general-purpose LLMs into audit-specific agents. LLMs are necessary but insufficient for audit work.
Native integrations with your audit methodology, engagement software, email, Excel, and more. The aOS operates as part of the firm's ecosystem, not alongside it.
The agent is constantly evolving, learning from current-year and prior-year workbooks in real time while referencing client emails and submitted source documents.
Audit-specific domain knowledge with reference to local and international standards such as IFRS and GAAP, ISA methodology, and Office and Google suite workflows.
Purpose-built surfaces for every audit workflow: drafting, research, preparation, client delivery, and workpaper finalisation. Built from the ground up for audit.
The foundation that makes the aOS enterprise-ready: ethical walls, cross-matter isolation, audit trails, and secure data storage according to your organisation's requirements.
How it works
You run the audit. Relay's agents do the execution — planning prep, evidence requests, population testing, exception analysis, and reviewer-ready write-ups — and bring every step back to you to sign off.
Please prepare an analytical procedure on the client's expenses.
Inspecting client's financial statements
Gaining understanding from dynamic knowledge base on the client
Checking materiality of the client
Inspecting audit methodology for sample collection
Creating workbook and selecting sample from the population
Awaiting review before sending to client
Audit Bench
Audit Bench is our AI evaluation framework for real-world audit work. It uses lifelike audit scenarios to benchmark how different models plan, inspect evidence, apply methodology, and produce reviewer-ready output.
Synthetic-but-grounded workpapers, source docs, populations, exceptions, and reviewer prompts.
Cases are shaped with practising audit judgement so models are tested against how audit work actually breaks.
Benchmark planning quality, evidence use, sampling decisions, exception handling, and final write-up quality.
FAQ
Relay works across the whole engagement rather than one workpaper at a time. When you're working in receivables, it already holds the bank statements, the client's emails about term sheets, the invoices and the open testing workbook to cross-reference. When the client's figures change, everything affected updates.
Relay keeps no client files. Source documents and drafted workbooks are never retained. What we do keep is an encrypted map of the engagement — file metadata, content summaries and how the documents relate to each other, plus limited account and security records. We keep it for as long as you're using Relay, so you don't have to re-upload each period, and we delete it when you stop using Relay or ask us to. Some AI processing happens outside South Africa; this is covered by POPIA's cross-border rules and set out in full in our privacy policy.
No, never. Every AI provider Relay uses is contractually bound to zero data retention: prompts and outputs are never stored after processing and are never used to train or improve any model.
Relay reads the files CaseWare produces on your computer — templates and workpapers included — so the two work together. Use both for as long as you like. Relay is built to replace legacy engagement software with something simpler and more powerful, but you can decide how it fits into your workflow.
Yes. Relay drafts workpapers from your own templates and methodology — your formatting, your referencing, your documentation standards — and applies them consistently across every engagement.
Audits are done in a fraction of the time, and quality goes up. Relay does the manual work; your team reviews, judges and signs.
Each engagement gets its own price, based on its size. The price is set before the audit starts and doesn't change with how it runs.
Request a demo below with your work email. We'll be in touch with your firm to arrange a time, and we'll walk through Relay on a real audit and talk about where it would fit your engagements.
Demo
Leave your work email and we'll be in touch with your firm to arrange a time.