Blab AI
Payward Accelerates Kraken Development with OpenAI Codex, Boosting Release Speed
When a $207 billion crypto exchange slashes its development cycle by half a year, the ripple is felt across the entire fintech ecosystem.
Payward, the parent company of Kraken, announced on Thursday that it has integrated OpenAI’s Codex coding assistant into its software pipeline. According to Kamo Asatryan, Chief Data Officer at Payward and Kraken, the AI tool has cut the company’s development lag by an estimated six months and has enabled a team of 50 agents to review merge requests simultaneously.
Kraken, founded in 2011, is a U.S.‑based exchange that was the first crypto firm to obtain a bank charter. By 2025 the platform had $207 billion in quarterly trading volume and was ranked the world’s fourteenth‑largest crypto exchange. In that same year, Kraken began offering tokenized equities to non‑U.S. customers, expanding its product suite beyond cryptocurrencies to include stocks, futures and ETFs.
Codex, released by OpenAI in 2023, is a large language model that translates natural language into executable code. It can run as a lightweight agent in a terminal, and recent updates have made it available on both macOS and Windows. The model is designed to assist developers with code generation, debugging, testing and documentation.
Asatryan explained that without Codex, Payward would likely be six months behind its current development schedule. He emphasized that speed is the team’s top priority, allowing the company to delegate routine coding tasks to the AI and focus human effort on higher‑level design and risk assessment. The 50‑agent setup enables parallel review of merge requests, reducing bottlenecks that traditionally slow release cycles.
Payward’s workforce is roughly 3,000 employees, according to a Bloomberg‑reported figure cited in a 2026 article. The same report noted that the company had recently cut about 150 positions after deploying AI tools, a move that coincided with a postponement of its initial public offering to 2027.
The integration of Codex is part of Payward’s broader strategy to streamline its technology stack. The company also offers a Kraken CLI that supports AI agents and global market access, allowing developers to programmatically place orders and manage trades through the exchange’s API.
Industry observers note that Payward’s adoption of Codex may influence other fintech and crypto firms to accelerate their own development pipelines. While the company has not released quantitative metrics on performance gains beyond the six‑month estimate, the move underscores the growing importance of AI‑assisted software development in high‑frequency, high‑volume trading environments.
OpenAI’s Codex has been used in a variety of contexts, from generating code for web applications to automating routine data‑engineering tasks. Its ability to run in parallel across multiple agents makes it well suited for environments where rapid iteration and continuous integration are critical.
Payward’s experience illustrates how AI tools can reduce development lead times without compromising code quality. By delegating routine coding tasks to Codex, the company can allocate human resources to more complex problem‑solving, potentially improving product reliability and security.
The company has not announced any immediate plans to release new products or features directly tied to Codex. However, the integration is expected to support upcoming updates to Kraken’s trading platform and its expanding suite of tokenized assets.
In summary, Payward’s use of OpenAI Codex represents a concrete example of AI accelerating software development in the financial sector. The reported six‑month speed‑up, combined with a 50‑agent review system, demonstrates how AI can streamline code review and release processes. As the fintech industry continues to adopt similar tools, the impact on development timelines and product quality is likely to grow.
The next steps for Payward will likely involve monitoring the long‑term effects of Codex on code quality, security, and developer productivity. The company’s experience may serve as a case study for other exchanges and financial technology firms seeking to leverage AI for faster, more efficient development cycles.
Payward, the parent company of Kraken, announced on Thursday that it has integrated OpenAI’s Codex coding assistant into its software pipeline. According to Kamo Asatryan, Chief Data Officer at Payward and Kraken, the AI tool has cut the company’s development lag by an estimated six months and has enabled a team of 50 agents to review merge requests simultaneously.
Kraken, founded in 2011, is a U.S.‑based exchange that was the first crypto firm to obtain a bank charter. By 2025 the platform had $207 billion in quarterly trading volume and was ranked the world’s fourteenth‑largest crypto exchange. In that same year, Kraken began offering tokenized equities to non‑U.S. customers, expanding its product suite beyond cryptocurrencies to include stocks, futures and ETFs.
Codex, released by OpenAI in 2023, is a large language model that translates natural language into executable code. It can run as a lightweight agent in a terminal, and recent updates have made it available on both macOS and Windows. The model is designed to assist developers with code generation, debugging, testing and documentation.
Asatryan explained that without Codex, Payward would likely be six months behind its current development schedule. He emphasized that speed is the team’s top priority, allowing the company to delegate routine coding tasks to the AI and focus human effort on higher‑level design and risk assessment. The 50‑agent setup enables parallel review of merge requests, reducing bottlenecks that traditionally slow release cycles.
Payward’s workforce is roughly 3,000 employees, according to a Bloomberg‑reported figure cited in a 2026 article. The same report noted that the company had recently cut about 150 positions after deploying AI tools, a move that coincided with a postponement of its initial public offering to 2027.
The integration of Codex is part of Payward’s broader strategy to streamline its technology stack. The company also offers a Kraken CLI that supports AI agents and global market access, allowing developers to programmatically place orders and manage trades through the exchange’s API.
Industry observers note that Payward’s adoption of Codex may influence other fintech and crypto firms to accelerate their own development pipelines. While the company has not released quantitative metrics on performance gains beyond the six‑month estimate, the move underscores the growing importance of AI‑assisted software development in high‑frequency, high‑volume trading environments.
OpenAI’s Codex has been used in a variety of contexts, from generating code for web applications to automating routine data‑engineering tasks. Its ability to run in parallel across multiple agents makes it well suited for environments where rapid iteration and continuous integration are critical.
Payward’s experience illustrates how AI tools can reduce development lead times without compromising code quality. By delegating routine coding tasks to Codex, the company can allocate human resources to more complex problem‑solving, potentially improving product reliability and security.
The company has not announced any immediate plans to release new products or features directly tied to Codex. However, the integration is expected to support upcoming updates to Kraken’s trading platform and its expanding suite of tokenized assets.
In summary, Payward’s use of OpenAI Codex represents a concrete example of AI accelerating software development in the financial sector. The reported six‑month speed‑up, combined with a 50‑agent review system, demonstrates how AI can streamline code review and release processes. As the fintech industry continues to adopt similar tools, the impact on development timelines and product quality is likely to grow.
The next steps for Payward will likely involve monitoring the long‑term effects of Codex on code quality, security, and developer productivity. The company’s experience may serve as a case study for other exchanges and financial technology firms seeking to leverage AI for faster, more efficient development cycles.
