About This Webinar
Join this workshop to explore AI classroom applications. This session will cover CodeHS's AI tools (Ask AI, AI Student Hints, Vibe Coding) and how to use AI in the classroom responsibly. Attendees will discuss practical strategies for integrating AI into instruction, so students stay engaged, accountable, and actually learning, not just prompting their way to an answer.
Explore CodeHS AI tools including Ask AI, Smart Hints, and Vibe Coding to support computer science instruction and foster learning in your classroom.
Full Transcript
Read the complete transcript of this webinar
I hope everyone's having a nice start to the school year. I know some people have probably been in school for a while now, while some school years are just starting. I'm based out of Michigan, and I know our schools just kind of got started either right before Labor Day or the week of, so we're kind of a late start here.
Well, as everyone's kind of getting logged in here, I just want to say thank you for joining me this evening or afternoon, depending on where you guys are at. I am Danielle Santelli, a PD specialist here at CodeHS. Before I started working here, I was a classroom teacher and taught K through 12, so quite the gamut. My last eight years were focused on computer science at the high school level. I did use CodeHS when I was in the classroom, and I only recently left—I've only been at CodeHS for about a year and a half. I was definitely in the classroom when generative AI hit the scene, so I know what the vibe is and how that's shaken things up.
I'm really excited to dig into our topic today, which is teaching with AI tools and strategies for the new school year. As you saw in our description, we're focusing on some of the tools we have within our platform. If you're new to CodeHS or if you've been around for a while, we have quite a lot of tools at your disposal, and some of them are AI-focused tools. We're going to focus on a couple of those today, dig into how we might want to use them in the classroom, explore the safeguards built into the platform to help you out when facilitating their use, and take a look at a few examples of things to look out for or strategies to use when utilizing these tools.
We're going to dig right in here. I already introduced myself, so we'll skip that slide. I want to share our AI approach here at CodeHS. Before we dig in, I want to talk about how CodeHS approaches AI as a company as we discuss our CodeHS AI tools.
First, districts are in control of those tools. Districts can decide which AI features are enabled or disabled based on their own policies. Second, even if your district allows it at the teacher level, you also have control and flexibility over your tool usage—what you want to allow or disallow. It's very easy: most of them are just toggles. You can toggle it on or off; nothing is permanent. You can decide to use it or not based on the day, the week, or what unit your students are working on—however you decide it works best for your classroom. Third, we have a lot of built-in safeguards. It's a very moderated AI, very limited in some of our capacities, and we're going to take a look at that as we dig into these tools today.
All right, so we are going to start with a little poll. Let me pull it up here. This is just a little check-in to see where we're at and gauge the temperature in the room regarding your students using AI.
The first option is if they're using it on their own and not in the classroom. Another option is having students use AI in your classroom on your terms, setting the expectation. Maybe you're just not sure—when they're not in your room, you can't really see what they're doing. You think they're being honest, but you never know. The last option is having a pretty strict policy about AI, or maybe your school has shut it down and you haven't incorporated it or found a way to fit it into what you're doing.
We'll take a moment to see where we're at. I know a couple of people joined while I was talking—thank you so much for joining us. We're digging into some of our AI tools on the CodeHS platform, exploring how they work alongside their moderation and control features.
We'll take another 30 seconds here. What's interesting is that we're kind of all over the board here; there isn't just one specific response. There is definitely a diverse set of ways our students are using AI. Before I left the classroom, I definitely had students using AI inappropriately. I taught web design, and a student copied and pasted straight from ChatGPT onto their assignment. It even included the little message saying, "I think this should fulfill your project requirements." I was like, wow, you couldn't even proofread!
I'm going to go ahead and end this poll so we can take a look. It looks like about half of us aren't really sure; maybe their students are smart enough to delete the ChatGPT messages when they submit. Hopefully, some of you have locked down restrictions to limit copy-paste capability. Some of you said you allow it in class on your terms, which is great for setting expectations. Others noted students use it outside for homework; if that's allowed, appropriate, and they're still understanding the content, there's nothing wrong with that. It's just about what we allow and what our students are doing. Thank you so much for engaging in that poll.
We're going to keep moving through here. One main idea we want to focus on is that AI can support learning, but students still need to do the thinking. That's always the big obstacle when discussing AI: we don't want to remove cognitive lift, instructional thinking, or student grit. How can we have the best of both worlds? How can we use a tool that is beneficial and helpful while still getting students to focus on learning and engaging with content?
Where we're spending most of our time today is looking at three AI tools:
1. AskAI, our teacher-facing AI tool.
2. Smart Hints, a brand-new student-facing AI tool.
3. Vibe Coding, which is for both students and teachers.
After looking at those, we'll spend time discussing how to use these tools in actual classrooms, view student usage data, and decide how to use that information to differentiate instruction or guide how we work with students.
If you are a CodeHS teacher and have an account, log in to follow along and play around with these tools if you can. Learning, listening, and doing at the same time is always the best way to learn, so I encourage you to log in and explore with me.
Let's start with Ask AI. Ask AI is our CodeHS teaching assistant, serving as a resource you can turn to for personalized information about your students and courses. Our tool can analyze the content you see and point you toward supplemental materials that work well for your specific courses. If you're an AP teacher, it can pull in AP resources to help you find relevant content.
If you are logged in to your teacher account and scroll way down, you'll see a new section in the navigation for this school year: the AI section. This is where a lot of our AI tools live. We'll select the second option, Ask AI.
It looks like a normal chatbot, featuring our AI, Carol. AI Carol wants to help. Down where you type your prompt, you'll notice options called personas. Personas guide what you want the AI to assist you with: general information, drafting emails, preparing materials, teaching tips, PD best practices, analyzing data, or finding course resources. These personas direct your prompts.
For example, if you want extra resources while teaching CSS, you can ask, "Some of my students are struggling with this—what resources could be helpful?" AI Carol searches through your courses and content to recommend handouts. If you ask for activities to teach Python loops, it digs through our project catalog to find options that fit your Python course. Ask AI generates active links directly in the chat, so you can click a recommended project to explore it, assign it to students, or continue the dialogue with the AI tool.
If you want to analyze how students are performing in a class, you can enable student data access by clicking the gear icon and allowing it to include student data. We care strongly about student data and ensure you remain in control. CodeHS takes integrity and transparency seriously: we do not share any data with third parties. Everything lives within the CodeHS universe. Even if you toggle that setting on, no data is shared outside CodeHS.
If you enable student data access and ask, "I have students struggling in web design—who needs additional intervention?", Ask AI looks at your sections, highlights who needs intervention, and can rank urgent needs. For instance, it might flag a failing student who flew under the radar. As a teacher-facing tool, only you have access to this, making it a versatile resource for finding content, managing assignments, and surfacing teaching resources.
You control what Ask AI sees. You must opt in through the gear icon to allow student data, and Ask AI will only summarize progress, grades, or plagiarism insights if enabled. Remember that Ask AI works like a flashlight: it highlights where to focus your attention, which is especially helpful when managing multiple classes of 30 students.
Here are a few prompt examples you can try:
- "Find a CodeHS lesson on loops that aligns with my state standards."
- "Which students haven't submitted a current assignment?"
- "Give me an extension project for students who finish early."
Feel free to put any questions in the Q&A or chat. Stephanie is behind the scenes navigating the chat. Stephanie also shared a link to our Ask AI Knowledge Base article, which is a great resource if you want to explore further.
Next, we'll cover Smart Hints, one of our student-facing tools. Smart Hints features limited AI interaction. When enabled, students working on an assignment have four basic choices:
1. "I'm not sure where to start" — AI Carol nudges them in the right direction.
2. "Give me a hint."
3. "Why is my code crashing?" — Helps when they can't pass test cases or autograders.
AI Carol reviews the code and points out specific lines or formatting issues without giving away the direct answer. It acts as a gentle nudge to look for typos or formatting errors. When staring at code or HTML, it's easy to miss minor typos, so this tool provides quick help.
On the student end, when they enter an assignment in a web design class and click the "Chat" option on the left, AI Carol pops up with those four choices. At the top, it displays "5 out of 5 hints left today." Students receive five hints total per day across all assignments and classes. Once spent, there is no way to reset or grant more hints.
If a student selects "I'm not sure where to start," AI Carol analyzes the code, explains relevant concepts (like style attributes), and suggests starting steps (like finding an H1 tag). It doesn't write the code or alter the screen; the student must read the advice and apply it themselves. Using a hint decrements their remaining daily count.
On the teacher end, you can access Smart Hints under the AI section. You can see which assignments students used hints on, student names, and the number of messages sent. If a student used multiple hints, you can click into the thread to review the conversation and check if they need intervention.
The Insights tab provides data on message volume per student and identifies assignments where hints are used most (e.g., boolean variables). You can use this data to determine what topics to review or reteach.
In the Smart Hints settings, the feature is default OFF. You must toggle it ON to grant student access. You can toggle access on or off for the whole class or individual students. If a student misuses hints, you can disable their access individually. You can turn it on for a challenging lesson and turn it off afterward, offering an initial layer of support before students raise their hands for help.
Our next tool is Vibe Coding, accessible to both students and teachers. Vibe coding allows users to build resources interactively with AI. You type a prompt (e.g., "Create an HTML page with specific instructions"), click go, iterate back and forth, and have the AI break down the code.
You can find Vibe Coding as the first option under the AI section. It displays prompt options alongside saved items you've built, such as an HTML tag matchup game or a Python loops quiz.
In Student Settings, you can enable student access to Vibe Coding for all students or individual students. Many teachers enable Vibe Coding only during class periods and turn it off afterward so students don't use up AI tokens outside of class. On the student end, Vibe Coding appears in their sandbox or left navigation, displaying their available credits without the administrative controls.
Teachers have full visibility into all student threads, allowing you to review their prompts and generated output. The tool features built-in moderation to flag off-topic or inappropriate requests. For example, if a student prompts, "Write my whole English essay on Great Expectations," AI Carol blocks the request. Flags also trigger if a student asks how to disable school monitoring software or requests medical advice. Blocked requests appear clearly with flag indicators and color coding.
This setup allows you to coach students on prompt engineering and iterative design. You hold the switch to toggle access on or off at any time. Note that Vibe Coding currently supports HTML projects only, does not support multi-user collaboration, and denies off-topic requests.
Teachers can also use Vibe Coding to prototype learning resources, such as an interactive HTML tag matching game. You can refine the code via prompts or edit the raw HTML directly in the interface. Once created, you can share the link with students or assign it to a course as an ungraded activity.
To review a few classroom scenarios:
Scenario 1: A student uses all five hints every day across all assignments.
This behavior suggests student dependency or a lack of AI literacy—viewing AI as a tool that completes work rather than one that assists learning. It signals a need to check in with the student to determine if they need content support or guidance on appropriate tool usage.
Scenario 2: A student prompts Vibe Coding with "Make it work now" and "Can you just write my history essay instead?"
This highlights the expectation of immediate results rather than engaging in iterative design. Vibe Coding relies on prompt engineering and step-by-step iteration. This scenario offers a teaching opportunity around writing detailed prompts and understanding that AI generation requires refining steps.
Scenario 3: A student submits a flawless project in 10 minutes but cannot explain a single function.
This illustrates the importance of focusing on process over product. Incorporating required checkpoints, documentation, and verbal explanations ensures students understand the concepts behind their work.
CodeHS provides support through our Knowledge Base, Ask AI tool, and the support icon at the bottom right of the screen. Additional resources include certified educator programs, micro-credentials, social media communities, feedback surveys, and upcoming webinars for AP Computer Science courses.
Well, as everyone's kind of getting logged in here, I just want to say thank you for joining me this evening or afternoon, depending on where you guys are at. I am Danielle Santelli, a PD specialist here at CodeHS. Before I started working here, I was a classroom teacher and taught K through 12, so quite the gamut. My last eight years were focused on computer science at the high school level. I did use CodeHS when I was in the classroom, and I only recently left—I've only been at CodeHS for about a year and a half. I was definitely in the classroom when generative AI hit the scene, so I know what the vibe is and how that's shaken things up.
I'm really excited to dig into our topic today, which is teaching with AI tools and strategies for the new school year. As you saw in our description, we're focusing on some of the tools we have within our platform. If you're new to CodeHS or if you've been around for a while, we have quite a lot of tools at your disposal, and some of them are AI-focused tools. We're going to focus on a couple of those today, dig into how we might want to use them in the classroom, explore the safeguards built into the platform to help you out when facilitating their use, and take a look at a few examples of things to look out for or strategies to use when utilizing these tools.
We're going to dig right in here. I already introduced myself, so we'll skip that slide. I want to share our AI approach here at CodeHS. Before we dig in, I want to talk about how CodeHS approaches AI as a company as we discuss our CodeHS AI tools.
First, districts are in control of those tools. Districts can decide which AI features are enabled or disabled based on their own policies. Second, even if your district allows it at the teacher level, you also have control and flexibility over your tool usage—what you want to allow or disallow. It's very easy: most of them are just toggles. You can toggle it on or off; nothing is permanent. You can decide to use it or not based on the day, the week, or what unit your students are working on—however you decide it works best for your classroom. Third, we have a lot of built-in safeguards. It's a very moderated AI, very limited in some of our capacities, and we're going to take a look at that as we dig into these tools today.
All right, so we are going to start with a little poll. Let me pull it up here. This is just a little check-in to see where we're at and gauge the temperature in the room regarding your students using AI.
The first option is if they're using it on their own and not in the classroom. Another option is having students use AI in your classroom on your terms, setting the expectation. Maybe you're just not sure—when they're not in your room, you can't really see what they're doing. You think they're being honest, but you never know. The last option is having a pretty strict policy about AI, or maybe your school has shut it down and you haven't incorporated it or found a way to fit it into what you're doing.
We'll take a moment to see where we're at. I know a couple of people joined while I was talking—thank you so much for joining us. We're digging into some of our AI tools on the CodeHS platform, exploring how they work alongside their moderation and control features.
We'll take another 30 seconds here. What's interesting is that we're kind of all over the board here; there isn't just one specific response. There is definitely a diverse set of ways our students are using AI. Before I left the classroom, I definitely had students using AI inappropriately. I taught web design, and a student copied and pasted straight from ChatGPT onto their assignment. It even included the little message saying, "I think this should fulfill your project requirements." I was like, wow, you couldn't even proofread!
I'm going to go ahead and end this poll so we can take a look. It looks like about half of us aren't really sure; maybe their students are smart enough to delete the ChatGPT messages when they submit. Hopefully, some of you have locked down restrictions to limit copy-paste capability. Some of you said you allow it in class on your terms, which is great for setting expectations. Others noted students use it outside for homework; if that's allowed, appropriate, and they're still understanding the content, there's nothing wrong with that. It's just about what we allow and what our students are doing. Thank you so much for engaging in that poll.
We're going to keep moving through here. One main idea we want to focus on is that AI can support learning, but students still need to do the thinking. That's always the big obstacle when discussing AI: we don't want to remove cognitive lift, instructional thinking, or student grit. How can we have the best of both worlds? How can we use a tool that is beneficial and helpful while still getting students to focus on learning and engaging with content?
Where we're spending most of our time today is looking at three AI tools:
1. AskAI, our teacher-facing AI tool.
2. Smart Hints, a brand-new student-facing AI tool.
3. Vibe Coding, which is for both students and teachers.
After looking at those, we'll spend time discussing how to use these tools in actual classrooms, view student usage data, and decide how to use that information to differentiate instruction or guide how we work with students.
If you are a CodeHS teacher and have an account, log in to follow along and play around with these tools if you can. Learning, listening, and doing at the same time is always the best way to learn, so I encourage you to log in and explore with me.
Let's start with Ask AI. Ask AI is our CodeHS teaching assistant, serving as a resource you can turn to for personalized information about your students and courses. Our tool can analyze the content you see and point you toward supplemental materials that work well for your specific courses. If you're an AP teacher, it can pull in AP resources to help you find relevant content.
If you are logged in to your teacher account and scroll way down, you'll see a new section in the navigation for this school year: the AI section. This is where a lot of our AI tools live. We'll select the second option, Ask AI.
It looks like a normal chatbot, featuring our AI, Carol. AI Carol wants to help. Down where you type your prompt, you'll notice options called personas. Personas guide what you want the AI to assist you with: general information, drafting emails, preparing materials, teaching tips, PD best practices, analyzing data, or finding course resources. These personas direct your prompts.
For example, if you want extra resources while teaching CSS, you can ask, "Some of my students are struggling with this—what resources could be helpful?" AI Carol searches through your courses and content to recommend handouts. If you ask for activities to teach Python loops, it digs through our project catalog to find options that fit your Python course. Ask AI generates active links directly in the chat, so you can click a recommended project to explore it, assign it to students, or continue the dialogue with the AI tool.
If you want to analyze how students are performing in a class, you can enable student data access by clicking the gear icon and allowing it to include student data. We care strongly about student data and ensure you remain in control. CodeHS takes integrity and transparency seriously: we do not share any data with third parties. Everything lives within the CodeHS universe. Even if you toggle that setting on, no data is shared outside CodeHS.
If you enable student data access and ask, "I have students struggling in web design—who needs additional intervention?", Ask AI looks at your sections, highlights who needs intervention, and can rank urgent needs. For instance, it might flag a failing student who flew under the radar. As a teacher-facing tool, only you have access to this, making it a versatile resource for finding content, managing assignments, and surfacing teaching resources.
You control what Ask AI sees. You must opt in through the gear icon to allow student data, and Ask AI will only summarize progress, grades, or plagiarism insights if enabled. Remember that Ask AI works like a flashlight: it highlights where to focus your attention, which is especially helpful when managing multiple classes of 30 students.
Here are a few prompt examples you can try:
- "Find a CodeHS lesson on loops that aligns with my state standards."
- "Which students haven't submitted a current assignment?"
- "Give me an extension project for students who finish early."
Feel free to put any questions in the Q&A or chat. Stephanie is behind the scenes navigating the chat. Stephanie also shared a link to our Ask AI Knowledge Base article, which is a great resource if you want to explore further.
Next, we'll cover Smart Hints, one of our student-facing tools. Smart Hints features limited AI interaction. When enabled, students working on an assignment have four basic choices:
1. "I'm not sure where to start" — AI Carol nudges them in the right direction.
2. "Give me a hint."
3. "Why is my code crashing?" — Helps when they can't pass test cases or autograders.
AI Carol reviews the code and points out specific lines or formatting issues without giving away the direct answer. It acts as a gentle nudge to look for typos or formatting errors. When staring at code or HTML, it's easy to miss minor typos, so this tool provides quick help.
On the student end, when they enter an assignment in a web design class and click the "Chat" option on the left, AI Carol pops up with those four choices. At the top, it displays "5 out of 5 hints left today." Students receive five hints total per day across all assignments and classes. Once spent, there is no way to reset or grant more hints.
If a student selects "I'm not sure where to start," AI Carol analyzes the code, explains relevant concepts (like style attributes), and suggests starting steps (like finding an H1 tag). It doesn't write the code or alter the screen; the student must read the advice and apply it themselves. Using a hint decrements their remaining daily count.
On the teacher end, you can access Smart Hints under the AI section. You can see which assignments students used hints on, student names, and the number of messages sent. If a student used multiple hints, you can click into the thread to review the conversation and check if they need intervention.
The Insights tab provides data on message volume per student and identifies assignments where hints are used most (e.g., boolean variables). You can use this data to determine what topics to review or reteach.
In the Smart Hints settings, the feature is default OFF. You must toggle it ON to grant student access. You can toggle access on or off for the whole class or individual students. If a student misuses hints, you can disable their access individually. You can turn it on for a challenging lesson and turn it off afterward, offering an initial layer of support before students raise their hands for help.
Our next tool is Vibe Coding, accessible to both students and teachers. Vibe coding allows users to build resources interactively with AI. You type a prompt (e.g., "Create an HTML page with specific instructions"), click go, iterate back and forth, and have the AI break down the code.
You can find Vibe Coding as the first option under the AI section. It displays prompt options alongside saved items you've built, such as an HTML tag matchup game or a Python loops quiz.
In Student Settings, you can enable student access to Vibe Coding for all students or individual students. Many teachers enable Vibe Coding only during class periods and turn it off afterward so students don't use up AI tokens outside of class. On the student end, Vibe Coding appears in their sandbox or left navigation, displaying their available credits without the administrative controls.
Teachers have full visibility into all student threads, allowing you to review their prompts and generated output. The tool features built-in moderation to flag off-topic or inappropriate requests. For example, if a student prompts, "Write my whole English essay on Great Expectations," AI Carol blocks the request. Flags also trigger if a student asks how to disable school monitoring software or requests medical advice. Blocked requests appear clearly with flag indicators and color coding.
This setup allows you to coach students on prompt engineering and iterative design. You hold the switch to toggle access on or off at any time. Note that Vibe Coding currently supports HTML projects only, does not support multi-user collaboration, and denies off-topic requests.
Teachers can also use Vibe Coding to prototype learning resources, such as an interactive HTML tag matching game. You can refine the code via prompts or edit the raw HTML directly in the interface. Once created, you can share the link with students or assign it to a course as an ungraded activity.
To review a few classroom scenarios:
Scenario 1: A student uses all five hints every day across all assignments.
This behavior suggests student dependency or a lack of AI literacy—viewing AI as a tool that completes work rather than one that assists learning. It signals a need to check in with the student to determine if they need content support or guidance on appropriate tool usage.
Scenario 2: A student prompts Vibe Coding with "Make it work now" and "Can you just write my history essay instead?"
This highlights the expectation of immediate results rather than engaging in iterative design. Vibe Coding relies on prompt engineering and step-by-step iteration. This scenario offers a teaching opportunity around writing detailed prompts and understanding that AI generation requires refining steps.
Scenario 3: A student submits a flawless project in 10 minutes but cannot explain a single function.
This illustrates the importance of focusing on process over product. Incorporating required checkpoints, documentation, and verbal explanations ensures students understand the concepts behind their work.
CodeHS provides support through our Knowledge Base, Ask AI tool, and the support icon at the bottom right of the screen. Additional resources include certified educator programs, micro-credentials, social media communities, feedback surveys, and upcoming webinars for AP Computer Science courses.
