Chatham Financial has integrated Codex and GPT‑5.6 into its technology stack to convert established processes into governed, repeatable workflows and software. By automating routine execution while preserving professional oversight, the firm aims to free experts to focus on higher‑value client work.
Process Zero: outcome-driven workflow redesign
Through its Process Zero reengineering service, Chatham restructures workflows around desired outcomes. For each workflow the firm identifies the minimum inputs and evidence required, determines where human judgment must remain, and decides how AI and AI-built tools should support the work. An early outcome of this approach is that trade validation time was reduced from about 30 minutes to under 4 minutes.
Trade validation and the Controls and Data Integrity team
Chatham’s Controls and Data Integrity team safeguards transaction data accuracy by validating that system records match what clients authorized and what was executed. Using Codex, the firm built a trade validation application that aggregates supporting transaction evidence, compares key terms, and flags discrepancies for review. Chatham is benchmarking the application’s results against experienced human reviewers and plans to extend it to more trade types while keeping controls and professional oversight in place.
Employee-built apps and Chatham Vibes
Chatham employees use ChatGPT and Codex in daily research, analysis, drafting, and software development. The internal Chatham Vibes platform enables staff to create applications tailored to their work. These apps are built with a range of models; AI features in Vibes apps run on GPT‑5.6 Terra by default, with GPT‑5.6 Sol available as an optional per‑app upgrade.
These internally developed applications increasingly support client-facing workflows: examples include tools to review maturing-cap trades and prepare pricing workbooks and client communications, as well as apps for producing fixed-income rate sheets, assembling hedging dashboards, and reviewing trade confirmations. Chatham professionals supply market context, client understanding, and judgment to evaluate and refine outputs before they reach clients.
Chatham Onyx: connected, auditable data and model routing
Chatham Onyx brings assets, debt, and derivatives into a connected environment where clients and advisors work from governed data and can use AI without losing traceability to source records. Codex is used across the Onyx development lifecycle to plan, build, test, document, and review software, shortening the journey from concept to durable product capability.
The Onyx platform leverages multiple OpenAI models — GPT‑5.6 Sol, GPT‑5.6 Terra, GPT‑5.4, and GPT‑4.1. Simpler analyses and non‑production testing are routed to more cost‑effective models, while GPT‑5.6 is used for complex tasks that require maximizing accuracy and value.
ChatFIN and structured insight
An example capability, ChatFIN, summarizes patterns in historical market data, helps users understand portfolios, and locates and links to legal documents for debt, derivative, and lease terms. These features help clients and advisors access information and insights, with Chatham’s advisors providing the subject matter expertise to interpret and apply those outputs.
Use of OpenAI for evidence comparison and next steps
Chatham uses OpenAI models to perform structured comparisons, organize evidence, identify exceptions, and let teams explore information more efficiently. This reduces the time advisors spend assembling information and increases time available for interpretation, difficult-case management, and client advice.
Next steps for Chatham include expanding trade validation to additional products, continuing to refine employee-built applications, and using Codex to develop new Onyx capabilities.
