pom.xml<\/code> file. If you haven\u2019t created a Spring Boot project yet, you can do so via Spring Initializr or manually in your IDE.<\/p>\n<dependencies>\n\t<dependency>\n\t\t<groupId>org.springframework.boot<\/groupId>\n\t\t<artifactId>spring-boot-starter-web<\/artifactId>\n\t<\/dependency>\n\t<dependency>\n\t\t<groupId>org.springframework.boot<\/groupId>\n\t\t<artifactId>spring-boot-starter-json<\/artifactId>\n\t<\/dependency>\n\t<dependency>\n\t\t<groupId>org.springframework<\/groupId>\n\t\t<artifactId>spring-web<\/artifactId>\n\t<\/dependency>\n\t<dependency>\n\t\t<groupId>org.springframework<\/groupId>\n\t\t<artifactId>spring-boot-starter-webflux<\/artifactId>\n\t<\/dependency>\n<\/dependencies>\n<\/pre>\n2.2 Setting Up application.properties<\/h3>\n
To keep our code clean, secure, and flexible, it’s a good practice to move configuration parameters, like the DeepSeek API Key and API endpoint URL, into the application.properties<\/code> file. This file is located in the src\/main\/resources\/<\/code> directory of your Spring Boot project.<\/p>\n# DeepSeek API configuration\ndeepseek.api.url=https:\/\/api.deepseek.com\/chatbot\ndeepseek.api.key=your-deepseek-api-key-here\n\n# Server configuration (optional)\nserver.port=8080\n<\/pre>\n2.3 Create the Service Layer<\/h3>\n
In this step, we will create a service class that will handle communication with the DeepSeek API. The service class will contain the logic to make HTTP requests to the DeepSeek API, send user input, and process the response.<\/p>\n
@Service\npublic class ChatbotService {\n\n\t@Value(\"${deepseek.api.url}\")\n\tprivate String apiUrl;\n\n\t@Value(\"${deepseek.api.key}\")\n\tprivate String apiKey;\n\n\t@Autowired\n\tprivate RestTemplate restTemplate;\n\n\tpublic String getResponseFromDeepSeek(String userMessage) {\n\t\t\/\/ Create request body\n\t\tMap<String, String> requestBody = new HashMap<>();\n\t\trequestBody.put(\"model\", \"deepseek-chat\");\n\t\trequestBody.put(\"message\", userMessage);\n\t\trequestBody.put(\"api_key\", apiKey);\n\n\t\tHttpHeaders headers = new HttpHeaders();\n\t\theaders.setContentType(MediaType.APPLICATION_JSON);\n\n\t\tHttpEntity<Map<String, String>> requestEntity = new HttpEntity<>(requestBody, headers);\n\n\t\t\/\/ Send POST request to DeepSeek API\n\t\tResponseEntity<String> response = restTemplate.exchange(\n\t\t\t\tapiUrl, HttpMethod.POST, requestEntity, String.class);\n\n\t\treturn response.getBody();\n\t}\n}\n<\/pre>\n2.3.1 Code Explanation<\/h4>\n
The ChatbotService<\/code> class is a Spring service component that communicates with the DeepSeek AI API. It uses the @Value<\/code> annotation to inject the DeepSeek API URL and API Key from the application.properties<\/code> file, promoting clean and configurable code. The class autowires a RestTemplate<\/code> to perform HTTP operations. Inside the getResponseFromDeepSeek<\/code> method, it constructs a request body as a map containing the user’s message and the API key, sets the appropriate HTTP headers with Content-Type: application\/json<\/code>, and then creates an HttpEntity<\/code> that wraps both the body and headers. It sends a POST request to the DeepSeek API using the RestTemplate.exchange()<\/code> method, passing the API URL, HTTP method, request entity, and expected response type. Finally, it returns the body of the API response, which typically contains the chatbot’s generated reply. This design cleanly separates configuration from logic, handles HTTP communication robustly, and follows Spring Boot best practices.<\/p>\n2.4 Controller to Handle User Input<\/h3>\n
Next, we\u2019ll create a REST controller to handle user input. This controller will accept user messages, pass them to the service layer, and return the chatbot\u2019s response.<\/p>\n
@RestController\n@RequestMapping(\"\/chat\")\npublic class ChatController {\n\n\t@Autowired\n\tprivate ChatbotService chatbotService;\n\n\t@PostMapping(\"\/sendMessage\")\n\tpublic String chat(@RequestBody String userMessage) {\n\t\treturn chatbotService.getResponseFromDeepSeek(userMessage);\n\t}\n}\n<\/pre>\n2.4.1 Code Explanation<\/h4>\n
The ChatController<\/code> class is a Spring REST controller that handles incoming HTTP requests for the chatbot application. It is annotated with @RestController<\/code> to indicate that it will handle RESTful web services and with @RequestMapping(\"\/chat\")<\/code> to prefix all its endpoints with “\/chat”. The controller injects an instance of ChatbotService<\/code> using @Autowired<\/code> for accessing the chatbot logic. The chat()<\/code> method is mapped to a POST endpoint “\/sendMessage” using @PostMapping<\/code>. It accepts the user’s message as the request body, passes this message to the getResponseFromDeepSeek()<\/code> method of the service layer, and returns the chatbot’s response directly as a plain String. This structure cleanly separates the controller from the service logic, making the code modular, readable, and maintainable.<\/p>\n2.5 Running the Application<\/h3>\n
Once you have completed the above steps, you can run your Spring Boot application. You can test it by sending POST requests to the \/chat\/sendMessage<\/code> endpoint. Below is an example of how to interact with the API using Postman or any other HTTP client:<\/p>\nPOST http:\/\/localhost:8080\/chat\/sendMessage\nContent-Type: application\/json\nBody:\n{\n\t\"message\": \"Hello, how are you?\"\n}\n<\/pre>\nUpon sending this request, you will receive a response generated by the DeepSeek AI model, such as:<\/p>\n
{\n\t\"response\": \"Hello! I'm an AI chatbot. How can I assist you today?\"\n}\n<\/pre>\n