Omio speeds product development by embedding OpenAI models

Omio has integrated OpenAI’s ChatGPT and Codex across its engineering operations, launched a conversational booking interface in 2023 tied to live transport data, and reports that technical effort to build products is now about 20% of previous levels. Projects that once required multiple developers for a quarter can now be completed by a single engineer in roughly one month.

The Prague-based multimodal travel platform coordinates services from more than 3,000 transportation providers across 47 countries. The company began by giving staff basic ChatGPT access to build familiarity with generative models before embedding OpenAI Codex directly into engineering workflows. Tomas Vocetka, Omio’s chief technology officer, required teams to redesign operational execution frameworks so processes would run with AI integrated at their core.

Codex is applied across the software development lifecycle: early research and architectural planning, active coding, automated testing, code reviews and maintenance. Engineering teams build internal connectors that link proprietary data stores and live inventory systems to the models, allowing developers to work inside their integrated development environments without pausing to fetch information. Vocetka described the initial ChatGPT rollout as a familiarisation phase and added that Codex now handles the bulk of production workload.

Internal analysis shows delivery timelines have shortened as development effort fell. Management reports that faster cycle times let teams prototype ideas quickly and validate consumer demand with minimal resources. The company reports it adjusts capital and engineering hours based on prototyping results to limit investment in features that fail early tests.

In 2023 Omio launched a conversational booking interface that connects OpenAI models to the company’s transportation inventory. The system accepts natural-language requests about complex multimodal routes, compares options such as flights versus trains and aggregates trains, buses, ferries and flights into unified, bookable itineraries. To prevent responses based on static training data, Omio grounds model outputs in live pricing and availability.

Omio expanded the setup into a dedicated ChatGPT experience that accesses its global transportation network. The application returns personalised journey options and serves as an interface layer between customers and the underlying inventory. The company describes the approach as conversational commerce, where the generative model mediates user intent and links it to booking systems to produce actionable results.

Codex use has been extended into non-technical corporate functions so standard operating procedures change alongside technical capabilities. Omio maintains a governance policy that keeps human personnel fully accountable for deployed code and final business outcomes. Tomas Vocetka, Omio’s CTO, emphasised, “The responsibility and accountability stay with people. AI helps us develop faster, analyse faster, and make decisions faster, but people stay in charge.”

The company’s oversight prevents automated systems from independently executing irreversible changes to booking infrastructure or core multimodal routing algorithms.