Automated web bot for flaying poker

The client provided me with an API service that, upon receiving information about the current state of the game table, returned the best recommended action. To develop the solution, I used Puppeteer.js as the main tool for browser automation. The bot was capable of automatically reading the state of tables across different online poker platforms in real time. It made decisions and placed bets automatically according to the recommendations from the API service. To evade anti-bot detection systems, the bot simulated realistic mouse movements and incorporated variable reaction times, introducing random delays into its actions to closely mimic human behavior. Additionally, I implemented an AI-based communication module that responded naturally to messages in the chat and strategically used emojis to influence the dynamics at the table. The bot was built with a modular architecture through the use of adapters: the core logic remained consistent, while specific adapters were developed for each poker website to handle interactions with their particular interfaces.