Can conductor's workers scale?

[rpe4a]

Hello! I need a help. For example, I just have a simple console java app with a poller and some workers inside of it.

TaskClient taskClient = new TaskClient();
taskClient.setRootURI("http://localhost:8080/api/");        //Point this to the server API

int threadCount = 2;            //number of threads used to execute workers.  To avoid starvation, should be same or more than number of workers

Worker worker1 = new SampleWorker("task_1");
Worker worker2 = new SampleWorker("task_5");

// Create TaskRunnerConfigurer
TaskRunnerConfigurer configurer = new TaskRunnerConfigurer.Builder(taskClient, Arrays.asList(worker1, worker2))
    .withThreadCount(threadCount)
    .build();

// Start the polling and execution of tasks
configurer.init();

I register 2 workers and start polling here. And now I have a question. What will happend if I run a second one yet?

They both start polling the server and at the same time they try to get some data from the same queue. Poller calls to the endpoint "poll/batch/{tasktype}" for data, but in the code futher I can't find anything which can prevent to get dublicate data or can synchronize this both calls. Especially here after a redis call you don't check data for duplicates. So in this case in the end the each worker will run the same task and we have a problem here.

1. May be am I wrong or miss something?
2. Can somebody explain how do you scale your workers in production?