DevRev(opens in a new window) built Computer and Operator to help service and operations teams carry even the hardest customer and employee requests through to completion. Computer and Operator are designed to reason across live data. Actions are reversible, and answers point back to their sources, explains Ahmed Bashir, DevRev’s chief technology officer.
DevRev combines OpenAI models, including GPT‑6 Astra and GPT‑6.1 Sol, with its harness for connecting company knowledge and giving teams control over the actions an agent can take. DevRev sees OpenAI models as providing reasoning and reliability with the speed and cost of serving customers and employees.
“It’s a combination of being able to have the right reasoning capabilities, being able to be trusted, highly aligned, and most importantly, it has to work from a latency perspective.”
—Ahmed Bashir, Chief Technology Officer, DevRev
Making AI work with company context
OpenAI models power the reasoning, planning, and tool selection behind both Computer and Operator. In DevRev’s Agent Studio, users can describe how they want a tool to behave in a conversation with AI and define parts of its behavior with fixed rules.
DevRev tests the models together with the software that supplies company context, enforces permissions, and controls actions. Its Enterprise-Bench benchmark evaluates how well they work when data has multiple versions, information sits in different systems, and access depends on a person’s role.
Ahmed says GPT‑6.1 Sol performs well in those evaluations. When a customer or employee is interacting with an AI agent, the model needs to understand the nuances of the request and respond quickly, at a cost the business can sustain. “I think it strikes the balance perfectly,” Ahmed says.
Building software and sales demos with Codex
Inside DevRev, engineers have made a clear choice about the best way to build. “Our entire team uses Codex,” Ahmed says. The adoption grew organically after engineers explored multiple coding tools. By the time of the interview, DevRev had moved fully to Codex and dropped its Cursor licenses.
For Ahmed, this reflects what developers value: “Developers are really just looking for the best tool, and you can’t fool them.” His engineers were unequivocal about wanting to continue with Codex.
“We think that Codex is far and away the best experience.”
—Ahmed Bashir, Chief Technology Officer, DevRev
DevRev’s sales engineers also use GPT‑6 Astra to create demos for the company’s booths at international conferences.
Responding faster to customers and employees
When a customer or employee asks for help, the request may require Operator to understand the issue, take approved action, or route to Computer to prepare the right person to act. Ahmed sees opportunities for the Decisions API and Ultrafast to help those workflows finish quickly, without requiring deep reasoning at every step.
Some of those decisions are difficult and expensive to define with fixed rules, especially when the information they depend on keeps changing. Ahmed wants builders to be able to describe more of that logic conversationally and choose the right kind of intelligence for each task. “These things have to be able to run and complete themselves in seconds, not minutes,” he says.
Realizing the promise of self-service
DevRev’s next ambition is for Operator to take on longer, more demanding customer and employee requests while keeping actions traceable, permissioned, and reversible.
Ahmed sees familiar support tasks, such as finding a knowledge-base answer, upgrading a plan, or getting a refund, as only the beginning. “We want to really realize that promise of self-service,” he says.
With OpenAI’s frontier intelligence, DevRev is building toward Computer and Operator carrying more complex requests through to resolution, with the company knowledge and controls to act on a person’s behalf.
Instagram Engineering
Shopify Engineering