AI at Confluence

Confluence POINT is AI-enabled automation that works within or alongside our solutions, delivering greater value to our clients.

Learn more

Confluence solutions are built on decades of knowledge and expertise, and firms have built their operations around them. They are specialist tools, and they work. Replacement is not on the roadmap. POINT extends these solutions to do more for our clients.

We understand that our technology automates a proportion of a task for our clients. We also know that manual effort still sits alongside that automation. Confluence POINT is shifting this balance through strategically applied AI. More automation, less manual effort. It is our new normal, with POINT we will optimize what already works.

AI at Confluence

Enhance

Optimize what already works

Our clients’ operations run on solutions built with decades of specialist knowledge. Firms are not looking to rip out high-performing and deeply-entrenched solutions. They want those systems to do more for them; to whittle down the manual time spent alongside this automation.

Foundation

Data is crucial

The data foundation that underpins AI success is crucial, and technology companies play a central role in undertaking the work needed to facilitate the AI rollout.

Flexibility

It’s not one size fits all

Not ever user wants to use a conversational interface. They become another way to get information from a system, but users will still need user interfaces that provide repeatable, auditable views that are powered by agents.

Transparency

The output from AI must be understood

Whether it’s regulatory reporting or performance measurement, output can never be “magic”. It must always be understood, traced back to proof points, validated with an audit trail. The risk tolerance here is low, and all AI development is with this in mind.

Product

POINT Validation

Financial firms handle a wide range of documents, often containing unstructured data, across a number of formats including PDF. This data is needed elsewhere in the business, and often these documents have to be cross-checked against each other. POINT Validation delivers AI-enabled document validation and intelligence, to ingest, transform, and validate documents.

When a firm produces a report using a technology solution, whether with Confluence or third party, it needs to be validated against data in a range of other documents and checklists before it is fully signed off. Historically this is where the role of structured technology ended, and manpower took over. This manual review could take 45+ minutes per document, scaling to thousands – even tens of thousands – of hours per year across a firm’s operations.

Learn more

How POINT Validation helps

POINT optimizes this workflow, delivering AI-powered validation. POINT can consume a wide range of document types and custom checklists, containing structured and unstructured data, which can be manually uploaded or flow from other systems. These could be PDFs, spreadsheets, documents, etc.
POINT then runs independent intelligent validation, checking across sources, ensuring data matches, and flagging inconsistencies. POINT validates a document within minutes.
Hands-on intervention is only needed to address the anomalies POINT identifies, and all findings come with an audit trail with workings clearly cited. POINT assists with routing anomalies via email or chat, and new actions or validation steps can be added on the fly – with the option to incorporate them for similar workflows in future.
POINT can also process the unstructured data it consumes to produce structured output for use in other workflows and systems. For example, it could take a PDF capital account statement and process it into a CSV file.

Product

POINT Prompt

Performance teams use Revolution for multi-asset attribution analysis of risk and performance but often switch back and forth between Revolution and Excel to achieve their goals. Confluence POINT has optimized this workflow via POINT Prompt, an AI-powered conversational interface for Revolution in Excel. You ask a question in plain English within Excel; POINT retrieves the answer from Revolution and returns it to you in Excel. No switching systems, no exporting, no navigation.

This could be the total level performance for the global equity portfolio year to date versus the benchmark; then to break down that performance by sector; then add pie chart showing asset allocation of the portfolio with some portfolio commentary. Requested in English within Excel; delivered by POINT, from Revolution to your worksheet.

Learn more

How POINT Prompt helps

POINT lets you ask in plain English, right where you work. It uses Model Context Protocol to give AI assistants like Claude, ChatGPT and Copilot a secure, direct line into Revolution’s performance and risk data. You work in Excel and type the question. No logging in, no navigating screens, no running reports.
POINT builds the answer one question at a time. Ask what is driving risk or how the portfolio performed against its benchmark, and the answer lands in your spreadsheet ready to work with. Every follow-up builds on the last: sector weights broken into asset allocation, security selection and total effect, a pie chart of allocation by GICS sector, portfolio commentary to go with it. Ask for a client-friendly one-page report and that arrives too, saved as XLS or PDF and ready for the meeting.
POINT delivers the same trusted numbers, far more accessible. It carries Revolution’s data and security, so the figures are exactly what you would expect from the platform itself. The insights are no longer locked behind a screen. They meet you where you already work, which means less time navigating and more time deciding.

AI Accelerators

The AI Accelerator series is an ongoing collection of conversations with people across Confluence who are at the forefront of how we build and use AI.

Nicholas Brien

Lead Software Engineer

In this installment, we talk to Nicholas Brien, Lead Software Engineer, who’s building AI into the products our clients use, while also leveraging it for a variety of tasks. Here’s where he thinks AI is genuinely changing his craft, and where he’s still skeptical. The AI Accelerator series is an ongoing collection of conversations with […]
Read more
Anas Cheriya

Principal Software Engineer

In this installment, we talk to Anas Cheriya, Principal Software Engineer, who has spent years building a hands-on relationship with AI, from early skepticism to architecting multi-agent systems in his own time. He’s embedded AI into the way his team develops software and is watching closely to see where the technology goes next. Here’s how his thinking has evolved, and where he still thinks humans need to stay […]
Read more
Chris Lombardi

Principal Software Engineer

In this installment of AI Accelerators, we sit down with Chris Lombardi, Principal Software Engineer, to explore how his relationship with AI evolved from cautious early experimentation to a core part of how he designs systems and makes engineering decisions. Chris shares how AI now shapes the products he builds for clients, the first tool […]
Read more
Stacey Nunez

Software Developer

In this installment, we speak to Stacey Nunez, a member of our software engineering team. Over four months, Stacey led the rebuild of an internal data tool for managing securities. She did this working solo alongside Anthropic’s Claude, demonstrating the leverage AI tools can offer. The result is a modern, fully tested solution that is in […]
Read more