Personalizing Your AI Tutor with System Prompts in C#

Personalizing Your Tutor with System Prompts

Welcome to the next step in your journey of creating a personal tutor with DeepSeek! In the previous lessons, you learned how to send messages to DeepSeek's language model, explored model parameters, and understood the importance of maintaining tutoring session history. Now, we will focus on personalizing your AI tutor by using system prompts . System prompts allow you to define the tutor's personality and expertise, enabling you to customize the behavior and tone of the AI's responses. This lesson will guide you through creating a specialized math and science tutor, demonstrating how system prompts can shape the tutor's interactions.

Creating a System Prompt

Join the 1M+ learners on CodeSignal

Be a part of our community of 1M+ users who develop and demonstrate their skills on CodeSignal

Talking to Your Specialized Tutor

To illustrate how a system prompt shapes the AI tutor's behavior, let's interact with an AI that has the math and science tutor personality set in our system prompt.

Below is a C# example that sends the session to the DeepSeek API and displays the tutor's response. This example uses a Message class to structure the conversation and a SendMessage method to handle the API call:

using System;
using System.Collections.Generic;
using System.Net.Http;
using System.Net.Http.Headers;
using System.Text;
using System.Text.Json;
using System.Threading.Tasks;

namespace PersonalTutor
{
    public class Message
    {
        public string role { get; set; }
        public string content { get; set; }

        public Message(string role, string content)
        {
            this.role = role;
            this.content = content;
        }
    }

    public class Program
    {
        public static async Task Main(string[] args)
        {
            var apiKey = Environment.GetEnvironmentVariable("OPENAI_API_KEY");
            var baseUri = Environment.GetEnvironmentVariable("OPENAI_BASE_URL");
            if (string.IsNullOrEmpty(apiKey))
            {
                Console.WriteLine("Set your OPENAI_API_KEY.");
                return;
            }
            if (string.IsNullOrEmpty(baseUri))
            {
                Console.WriteLine("Set your OPENAI_BASE_URL (e.g. https://api.openai.com).");
                return;
            }

            var endpoint = baseUri.TrimEnd('/') + "/v1/chat/completions";

            // Define the tutor's personality and expertise
            string systemPrompt = "You are an experienced and patient tutor specialized in math and science, providing clear and concise explanations.";

            // Structure the tutoring session with a system prompt and student query
            var session = new List<Message>
            {
                new Message("system", systemPrompt),
                new Message("user", "Can you help me understand the concept of derivatives in calculus?")
            };

            // Send the session and get the AI's response
            string reply = await SendMessage(endpoint, apiKey, session);
            Console.WriteLine("Tutor Response: " + reply);
        }

        private static async Task<string> SendMessage(string endpoint, string apiKey, List<Message> conversation)
        {
            using var client = new HttpClient();
            client.DefaultRequestHeaders.Authorization = new AuthenticationHeaderValue("Bearer", apiKey);

            var payload = new
            {
                model = "deepseek-ai/DeepSeek-V3",
                messages = conversation
            };

            var jsonPayload = JsonSerializer.Serialize(payload);
            using var content = new StringContent(jsonPayload, Encoding.UTF8, "application/json");
            var response = await client.PostAsync(endpoint, content);
            var body = await response.Content.ReadAsStringAsync();

            if (!response.IsSuccessStatusCode)
            {
                Console.WriteLine($"Error ({response.StatusCode}): {body}");
                return "";
            }

            using var doc = JsonDocument.Parse(body);
            var reply = doc.RootElement
                           .GetProperty("choices")[0]
                           .GetProperty("message")
                           .GetProperty("content")
                           .GetString();

            return reply?.Trim() ?? "";
        }
    }
}

using System;
using System.Collections.Generic;
using System.Net.Http;
using System.Net.Http.Headers;
using System.Text;
using System.Text.Json;
using System.Threading.Tasks;

namespace PersonalTutor
{
    public class Message
    {
        public string role { get; set; }
        public string content { get; set; }

        public Message(string role, string content)
        {
            this.role = role;
            this.content = content;
        }
    }

    public class Program
    {
        public static async Task Main(string[] args)
        {
            var apiKey = Environment.GetEnvironmentVariable("OPENAI_API_KEY");
            var baseUri = Environment.GetEnvironmentVariable("OPENAI_BASE_URL");
            if (string.IsNullOrEmpty(apiKey))
            {
                Console.WriteLine("Set your OPENAI_API_KEY.");
                return;
            }
            if (string.IsNullOrEmpty(baseUri))
            {
                Console.WriteLine("Set your OPENAI_BASE_URL (e.g. https://api.openai.com).");
                return;
            }

            var endpoint = baseUri.TrimEnd('/') + "/v1/chat/completions";

            // Define the tutor's personality and expertise
            string systemPrompt = "You are an experienced and patient tutor specialized in math and science, providing clear and concise explanations.";

            // Structure the tutoring session with a system prompt and student query
            var session = new List<Message>
            {
                new Message("system", systemPrompt),
                new Message("user", "Can you help me understand the concept of derivatives in calculus?")
            };

            // Send the session and get the AI's response
            string reply = await SendMessage(endpoint, apiKey, session);
            Console.WriteLine("Tutor Response: " + reply);
        }

        private static async Task<string> SendMessage(string endpoint, string apiKey, List<Message> conversation)
        {
            using var client = new HttpClient();
            client.DefaultRequestHeaders.Authorization = new AuthenticationHeaderValue("Bearer", apiKey);

            var payload = new
            {
                model = "deepseek-ai/DeepSeek-V3",
                messages = conversation
            };

            var jsonPayload = JsonSerializer.Serialize(payload);
            using var content = new StringContent(jsonPayload, Encoding.UTF8, "application/json");
            var response = await client.PostAsync(endpoint, content);
            var body = await response.Content.ReadAsStringAsync();

            if (!response.IsSuccessStatusCode)
            {
                Console.WriteLine($"Error ({response.StatusCode}): {body}");
                return "";
            }

            using var doc = JsonDocument.Parse(body);
            var reply = doc.RootElement
                           .GetProperty("choices")[0]
                           .GetProperty("message")
                           .GetProperty("content")
                           .GetString();

            return reply?.Trim() ?? "";
        }
    }
}

When you run this code, you will receive a response from the AI tutor that reflects the experienced math and science tutor personality. The output might look something like this:

text

Tutor Response: Derivatives in calculus represent the rate of change of a function at a specific point. Think of it as finding the slope of the tangent line to a curve at any given point.

The derivative of a function f(x) is written as f'(x) or df/dx, and it tells us how quickly the output value changes as we slightly adjust the input value.

For example, if f(x) = x², then f'(x) = 2x. This means that at any point x, the rate of change is 2x. At x = 3, the derivative equals 6, indicating the function is increasing at a rate of 6 units per unit change in x at that point.

Would you like me to explain how to calculate derivatives using specific rules or provide more examples?

Tutor Response: Derivatives in calculus represent the rate of change of a function at a specific point. Think of it as finding the slope of the tangent line to a curve at any given point.

The derivative of a function f(x) is written as f'(x) or df/dx, and it tells us how quickly the output value changes as we slightly adjust the input value.

For example, if f(x) = x², then f'(x) = 2x. This means that at any point x, the rate of change is 2x. At x = 3, the derivative equals 6, indicating the function is increasing at a rate of 6 units per unit change in x at that point.

Would you like me to explain how to calculate derivatives using specific rules or provide more examples?

This educational response showcases how the system prompt shapes the tutor's interactions, allowing it to respond with clear, concise explanations focused on mathematical concepts — precisely what we defined in our system prompt.

Customizing Tutor Expertise

One of the powerful aspects of system prompts is that they can be tailored to create tutors with different specializations or teaching styles. Let's look at how we might create a tutor with a different focus:

// Define a different tutor specialization
string historyTutorPrompt = "You are a passionate history tutor who specializes in making historical events engaging and memorable through storytelling and interesting facts.";

// Structure a new tutoring session
var historySession = new List<Message>
{
    new Message("system", historyTutorPrompt),
    new Message("user", "Can you tell me about the Industrial Revolution?")
};

// Get response from the history-focused tutor
string historyReply = await SendMessage(endpoint, apiKey, historySession);
Console.WriteLine("History Tutor Response: " + historyReply);

// Define a different tutor specialization
string historyTutorPrompt = "You are a passionate history tutor who specializes in making historical events engaging and memorable through storytelling and interesting facts.";

// Structure a new tutoring session
var historySession = new List<Message>
{
    new Message("system", historyTutorPrompt),
    new Message("user", "Can you tell me about the Industrial Revolution?")
};

// Get response from the history-focused tutor
string historyReply = await SendMessage(endpoint, apiKey, historySession);
Console.WriteLine("History Tutor Response: " + historyReply);

Running this code results in an output similar to the following: