Skip to content
Main Site News Console

Merge Builds Products and Business Tools with OpenAI

· OpenAI Translated
OpenAILLM

Merge Builds Products and Business Tools with OpenAI

November 5, 2026

The integration platform uses GPT‑6 Astra, ChatGPT, and Codex for product code reviews, marketing, and business development planning.

Image 1: The white Merge logo centered on a purple brushed-texture background.

Company size: Startup
Region: North America
Industry: Technology
Products used: ChatGPT, Codex

Merge helps companies build interconnected software and AI products, spanning capabilities from syncing customer data to enabling AI agents to take action across business systems. Its product portfolio includes: customer data integration through Merge Unified, secure tool-calling access for AI agents via Merge Agent Handler, and model routing and cost controls through Merge Gateway. Merge for Workforce brings these capabilities internally, helping teams manage the models, tools, and business systems employees can access. Working together, these products empower both the external software companies build for customers and the AI-driven operations that power their internal businesses.

Internally at Merge, CEO and Co-Founder Shensi Ding uses Codex to review product code, including work on a new settings interface for configuring tool access in Merge for Workforce. Her colleagues use GPT‑6 Astra for product testing and marketing. The team also applies AI to day-to-day operations, from preparing sales pipeline reviews to hosting customer dinners.

“Honestly, we use AI for almost everything. Beyond data integration, we use it to automate workflows, improve product quality across the board, synthesize customer feedback, and build end-to-end features.”

—Shensi Ding, CEO and Co-Founder, Merge

In Practice

Merge launched a tool for adding connectors to AI products, and Ding puts reliability first: “We take immense pride in our product quality. Tool calls made through Merge have a 99.9% success rate, compared to roughly 60% for public MCP (Model Context Protocol) servers.”

In its own product development, Merge uses GPT‑6 Astra. “It’s blazing fast, and its performance on code reviews is dramatically improved,” says Ding. “People also love GPT‑6 Astra’s personality—it’s really engaging. At the same time, Codex’s progress is undeniable. Merge lets employees choose their own AI tools, and which one to use has become a lively internal debate.”

When the team evaluated how customers would configure tool access, Ding used Codex to review the proposed implementation of the new Provisioning Tab in Merge for Workforce. They discussed designing the “Add tool pack” action as a clearer standalone open link to lower the cognitive overhead of the setup flow. Codex provided code review feedback; the Merge team made product decisions and adjusted the implementation accordingly, with engineers handling final validation and shipping.

“Other AI systems tend to forget what I said earlier and only follow my last instruction. GPT-6 Astra consistently executes every task exceptionally well and proactively fills in details I didn’t explicitly spell out, using common sense.”

—Shensi Ding, CEO and Co-Founder, Merge

The Dinner Bot

Merge hosts executive dinners for current and prospective customers to forge connections and create opportunities for downstream product conversations.

For a dinner held in Dallas, Merge identified the city and the target guest profile. The Dinner Bot then pulled potential attendee info from CRM and Apollo, cross-referenced past invitations and RSVPs, and helped draft outreach emails sent from Ding’s inbox. In addition, it maintained attendee tracking sheets, suggested seating arrangements, and generated calendar notes so sales and Customer Success (CS) teams could add background context for each guest.

The model handled reasoning and drafting, while Merge’s connectors provided access to underlying business systems. Ding served as the human-in-the-loop, polishing any unnatural phrasing and curating the final guest list using personal context the bot lacked.

In October 2026, Merge reported that the effort required to build guest lists, send outreach emails, and maintain RSVP trackers dropped from dozens of hours across the company to about 30 minutes per event. This comparison was measured against identical planning tasks over the prior three months. The team now focuses solely on hosting, relationship building, and on-the-ground logistics.

Merge hosted 16 executive dinners in Q3—double the 8 held in Q2—and plans to host more than 30 in Q4. Following the Dallas dinner, the team received positive responses to the invitations and secured opportunities for follow-up product demos.

“With GPT‑6 Astra, the Dinner Bot connects to our Salesforce, Gong, Google Sheets, and Gmail, sending emails to people in our network on my behalf,” says Ding. “It suggests the right people for each dinner and surfaces prospective attendees we hadn’t thought of.”

Everyone Is a Builder

Ding notes: “It’s especially empowering for non-technical team members, because now they can build whatever they want.”

Charlie Lynch, a sales leader at Merge, built reusable AI workflows for deal approvals, forecasting updates, and post-renewal follow-ups. Instead of repeatedly hunting down context for every update or decision, he relies on workflows to prepare materials in advance, leaving him to simply review recommendations and decide on next steps.

The social media team uses GPT‑6 Astra in ChatGPT to identify posts the company should engage with. The marketing team also leveraged it to refine and clarify product messaging for the launch of Merge Gateway Evaluations.

Leadership Insights

  • Lead from the front and use the tools yourself: “If you’re an executive, your understanding sets the ceiling for AI proficiency across the entire company,” Ding says.
  • Find low-stakes scenarios to experiment: Ding and her co-founder chose to explore AI on a new product before customers were reliant on the codebase. “When you’re less anxious and free to experiment, you learn so much more,” she explains.
  • Give employees time and space to explore: Merge lets every team member choose their preferred tools and encourages them to spend time learning and testing every day. That includes giving them the freedom to “move fast and make mistakes,” even if it means pursuing a dead end that yields no tangible output.

Looking Ahead

“Whatever I want to do, I just tell the AI, and it gets done immediately,” Ding says. Merge continues to push the boundaries of what its team can build using OpenAI’s technology, bringing the same spirit of experimentation to product development and business operations alike.