Blog post
Beyond the AI Hype: How Vyking Teams Use AI in Everyday Work
“AI at Vyking did not start as a top-down transformation project. It started where useful technology usually starts: inside real workflows, solving real problems.”
Across our engineering, operations, DevOps, intelligence, and frontend teams, AI has gradually become part of the way people work. It helps teams debug issues, write documentation, analyze logs, validate queries, speed up integrations, improve communication, and reduce repetitive manual effort. The aim was never to replace expertise. It was to give expert teams more time for the work that matters most: architecture, product quality, operational thinking, platform stability, and better player experiences.
What began as small experiments with coding assistance, documentation, and research has now become part of everyday workflows across the business. AI is not treated as a standalone trend at Vyking. It is becoming a practical layer inside the way teams build, support, and improve modern iGaming technology.
Why AI Matters in iGaming Platform Development
Building and operating an iGaming platform involves a lot of moving parts.
Provider integrations, payment flows, localization requirements, regulatory differences, frontend journeys, player account systems, reporting, and operational support all create constant technical and operational complexity. For Vyking, AI is useful because it helps teams move through that complexity faster. It supports quicker investigation, clearer documentation, more consistent testing, faster understanding of unfamiliar systems, and better organization of information across technical and operational workflows.
Most importantly, it helps reduce the friction that builds up in busy teams. As repetitive work starts to shrink, people have more space to focus on problem-solving, communication, product quality, and the decisions that require real platform context.
Which AI Tools Do Teams Use?
Different Vyking teams use different AI tools depending on the type of work involved.
Commonly used tools include:
General AI assistants: ChatGPT, Claude, Gemini Deep Research
Engineering tools: GitHub Copilot, Claude Code, Codex, Cursor, MCP tools
Workflow and content tools: ClickUp AI, AI Agents, Gamma
Rather than standardizing around one model or one way of working, teams choose the tool that best fits the problem in front of them. Some engineers prefer terminal-based AI agents that can work directly inside large codebases. Others combine different tools depending on whether they are debugging, researching, documenting, brainstorming, or reviewing technical decisions. Operations teams often use AI to organize information and improve reporting. DevOps teams lean on it for troubleshooting, prototyping, infrastructure discussions, and automation planning. Over time, usage has moved far beyond simple prompts or quick searches.
Today, teams regularly use AI for documentation, debugging, coding, log analysis, SQL validation, support ticket analysis, brainstorming, translations, reporting, testing, troubleshooting, communication improvements, operational research, UI implementation from design systems, localization workflows, API investigation, and frontend development acceleration.
One of the most noticeable changes is simple: repetitive work has started to shrink.
Tasks like generating boilerplate code, creating test structures, summarizing incidents, organizing operational notes, reviewing large log files, and preparing documentation now take far less manual effort than before. Several teams also mentioned that AI has lowered the “automation barrier.” Small utility scripts, internal tools, and one-off automations that previously felt too time-consuming to build are now created more often because the effort required is much lower. Documentation came up again and again. Not because teams did not value it, but because in busy delivery cycles, implementation usually came first. With AI-assisted workflows, creating and improving documentation has become easier, faster, and more consistent.
What Improvements Have Teams Seen?
The impact varies by team and by task, but the same patterns appeared across different departments.
Engineering teams described debugging and troubleshooting as roughly twice as fast in many cases. Repetitive scripting, log analysis, and test generation often saw even larger improvements, especially when AI was used to speed up investigation or produce repeatable structures. In some existing codebases, teams reported that certain feature delivery and integration workflows could move several times faster than before. The biggest gains appeared where the work involved repeated patterns, boilerplate, API investigation, or documentation-heavy implementation.
DevOps teams estimated that selected infrastructure, troubleshooting, and automation workflows became around 2x–2.5x faster after integrating AI into daily work. For Operations, the gains showed up slightly differently. The value was less about pure delivery speed and more about clarity, consistency, and reduced manual effort when summarizing incidents, organizing tickets, preparing reports, and reviewing fragmented operational information.
Teams also pointed to improvements that are harder to measure but just as important:
better documentation
broader test coverage
clearer communication
more consistent processes
faster investigation
improved reporting quality
more time for architecture and decision-making
One point came through clearly across teams: AI does not only make existing work faster. It also makes teams more likely to complete the tasks that usually fall behind when things get busy. Documentation, testing, refactoring, communication clean-up, and small automations become easier to maintain as part of everyday work.
AI in the Business Intelligence Team
Inside the Vyking Intelligence team, AI has become part of how engineers investigate issues, write documentation, test functionality, and reduce repetitive implementation work. The team regularly uses tools such as Claude, ChatGPT, MCP tools, and Gamma for coding, debugging, research, brainstorming, translations, documentation, and testing.
A major shift came in troubleshooting and data analysis. Tasks that previously required long periods of manual investigation can now often be narrowed down much faster with AI assistance. Complex debugging sessions and repetitive investigation work have become easier to manage compared with older workflows. The team also highlighted strong improvements in documentation quality and testing coverage. AI helps generate repetitive structures such as route handlers, schemas, fixtures, and larger testing scenarios, allowing engineers to focus more on logic, architecture, and platform behaviour instead of repetitive implementation.
For the Intelligence team, the value is not just faster output. It is the ability to maintain better technical hygiene while still moving quickly.
AI in the Operations Team
For the Operations team, AI quickly became a practical productivity tool used throughout daily work. Teams commonly use Claude, ChatGPT, ClickUp AI, AI Agents, and Copilot depending on the task. AI is used for documentation, incident summaries, troubleshooting, SQL validation, reporting, and organizing large amounts of operational information into structured updates. One of the clearest benefits has been speed and clarity during incident handling.
Operational work often means reviewing fragmented tickets, logs, comments, and updates from multiple systems. AI helps turn that information into summaries, timelines, and clearer internal updates much faster than manual processing alone. The team also uses AI during investigations and research. It can help identify patterns, suggest additional troubleshooting paths, or explain unfamiliar concepts while teams are actively working through a problem.
For Operations, AI is most useful when it brings order to complexity. It helps teams move through large amounts of information faster and communicate more clearly across technical and non-technical stakeholders.
AI in the DevOps Team
Within the DevOps team, AI adoption has focused heavily on automation, troubleshooting, infrastructure planning, and faster investigation. The team mainly uses Claude Code, Codex, and Gemini Deep Research for prototyping, repetitive implementation tasks, documentation generation, research, and internal automation work.
Log analysis has become one of the most practical use cases. Large access logs that previously required significant manual investigation can now be reviewed much faster with AI assistance. In some cases, AI is simply more efficient at navigating large log datasets and surfacing suspicious patterns quickly. The team also uses AI when discussing infrastructure improvements, identifying bottlenecks, evaluating scalability questions, and planning internal system optimizations.
Several DevOps workflows are now estimated to be roughly 2x–2.5x faster after integrating AI into daily work. The strongest gains appear in repetitive implementation, investigation, research, and documentation-heavy tasks. For DevOps, AI helps reduce manual load while improving the speed of technical exploration and infrastructure planning.
AI in Engineering/Dev Team
Engineering teams across Vyking have also started integrating AI more deeply into development workflows.In some projects, AI-assisted frameworks are now used to accelerate provider integrations, automate testing, improve documentation quality, and simplify repetitive engineering tasks.
Processes that previously required several weeks of repetitive implementation work can now often be completed dramatically faster with structured AI-assisted workflows. Teams also highlighted an important qualitative change: documentation, testing, and refactoring are no longer constantly pushed aside because the effort required to produce them has become significantly lower.
That matters for maintainability. Faster delivery is valuable, but the larger benefit is that teams can more consistently support the practices that make systems easier to understand, improve, and scale over time. For engineering teams, AI reduces the amount of time spent on repetitive implementation so more attention can go toward architecture, logic, product quality, and long-term platform health.
AI in the Frontend Team
AI has become a regular part of frontend development workflows across player-facing areas of the platform, including cashier, sportsbook, authentication, and other core user journeys. One of the most valuable improvements came through the use of AI together with Figma integrations and MCP servers. By allowing AI to access design context directly, including layouts, spacing, components, and design tokens, frontend engineers can reduce the amount of manual translation between design and implementation.
This helps close the gap between design and code while accelerating feature delivery. AI is also frequently used during API investigations and integration work. Developers use it to trace responses, validate expected behaviour, and navigate complex frontend-backend interactions more efficiently.
Another common use case involves localization management, where AI helps maintain formatting consistency across large sets of translated content while keeping translations aligned with UI requirements. Like other engineering teams, the biggest gains come from removing repetitive steps. Faster scaffolding, quicker investigations, and smoother implementation workflows allow engineers to spend more time focusing on user experience, application architecture, and product quality.
For the Frontend team, the value is clear: faster implementation, fewer repetitive handovers, and more consistency across the player-facing journeys that matter most.
A Culture of Practical AI Adoption
One of the reasons AI adoption has worked well at Vyking is that it grew from practical experimentation. Different teams discovered different workflows, tools, and approaches based on their own challenges. A useful prompt, agent setup, documentation workflow, debugging method, or reporting structure found by one team could often help another. Over time, AI adoption became less about isolated experimentation and more about shared operational learning.
That internal knowledge sharing has been important. It helps teams adopt useful practices faster, avoid repeating the same experiments, and build a more practical understanding of where AI genuinely adds value. The bottom-up nature of adoption also made it more realistic. Teams were not using AI because it was fashionable. They were using it because it helped them solve real problems faster.
Final Thoughts
Across Vyking, the clearest lesson is that AI works best when it is practical, specific, and connected to real workflows. It helps teams move faster through repetitive tasks, improve documentation, summarize complex information, accelerate troubleshooting, and create more room for higher-value thinking.
But the real value is not simply speed.It is consistency.
Better documentation becomes easier to maintain. Broader testing becomes easier to produce. Incident communication becomes clearer. Engineering teams can spend more time on architecture and quality. Operations teams can move faster through complexity. DevOps teams can investigate, prototype, and plan infrastructure improvements more efficiently. That is where AI is already making a difference at Vyking: not as a standalone trend, but as a practical layer inside the way our teams build, support, and improve modern iGaming technology.
