DEV Community: T Frazao

Advanced Code & Security Analyzer AI Agent

T Frazao — Mon, 27 Jan 2025 00:15:05 +0000

This is a submission for the Agent.ai Challenge: Full-Stack Agent (See Details)

This is a submission for the Agent.ai Challenge: Productivity-Pro Agent (See Details)

What I Built

The Advanced Code & Security Analyzer was born from a critical observation: as AI agents increasingly handle sensitive user data, there's an urgent need to ensure robust security practices in their implementation. I built this agent to help developers be more mindful of security vulnerabilities that could compromise user data protection.

Motivation:

Growing concerns about AI agents' access to sensitive user information
Need for automated security checks in AI-driven development
Desire to promote security-first thinking in the AI agent ecosystem
Recognition that many data breaches stem from basic security oversights

Key Features:

1 Flexible Input Methods: Accepts code via file uploads or GitHub repositories
2 Multi-Stage Analysis:

Initial code review and language detection
Comprehensive security vulnerability scanning
GDPR compliance assessment
Severity classification (Critical/Moderate/Low)

3 Detailed Reporting:

Executive summary of findings
Detailed vulnerability analysis with risk levels
OWASP Top 10 mapping
Actionable recommendations
Citations and references

4 Automated Alerts: Sends urgent notifications for identified security issues with detailed markdown-formatted reports

The agent is particularly valuable for:

Teams developing AI agents handling user data
Development teams needing quick security assessments
Security teams performing code audits
Organizations requiring GDPR compliance checks
Projects using environment variables and API keys

Use Case Workflow:

1 Developer uploads code for security analysis
2 Agent performs comprehensive security audit
3 Based on severity findings:

CRITICAL: Sends urgent email alert to developer and creates a report
MODERATE/LOW: Creates report for documentation Developer can access:
Immediate email notification with key findings
Detailed Doc report for thorough review and sharing
Actionable recommendations for security improvements

The agent serves as both a security tool and an educational resource, helping developers understand and implement better security practices in their AI agent development process. The dual-delivery system (email + Document) ensures that critical security issues are immediately addressed while maintaining a permanent record of all security audits for future reference and compliance purposes.

This automated workflow helps teams:

Respond quickly to critical security issues
Maintain audit trails of security reviews
Share findings easily with stakeholders
Track security improvements over time
Build more secure AI agents from the ground up

Demo

Link: https://agent.ai/agent/CodeSecurityAudit

Link to video: https://youtu.be/961h9P0rf8s

Screenshots

Agent.ai Experience

The overall experience was good. I appreciated the informative notes about built-in functions, the clear user interface, and its user-friendly nature. Nothing was too complex. However, this simplicity limits the agent's customization capabilities.

In my experience, I encountered significant difficulties working with webhooks. I developed a VS Code extension that used webhooks to analyze file content while developers were working and then opened a new VS window with the analysis report. After numerous attempts and troubleshooting, even the basic curl command for the webhook wasn't functioning properly, consistently returning the error: "Error running agent: cannot unpack non-iterable NoneType object."

After several attempts using only the default curl command, the same error persisted. As a result, I put that feature of the agent on hold. Additionally, besides using email, I tried the Agent.ai built-in function to save a document to Google Drive, but it failed to work as smoothly as the email function. I just wished debugging could be more developer friendly and also more documentation for more complex workflows, for example, what data format the agent is expecting to receive through the webhook.

In conclusion, the platform has significant potential, and the community will likely grow rapidly. Despite the challenges, it was an intense but rewarding experience. Thank you.

Reddit Launches 'Answers': A New AI-Powered Feature to Surface Community Knowledge

T Frazao — Wed, 22 Jan 2025 00:07:24 +0000

Reddit, the front page of the internet, has stepped into the generative AI arena with its latest feature called "Answers." This innovative tool represents a significant shift in how users can access and digest the vast knowledge base within Reddit's communities. Let's dive into what this means for both casual browsers and power users.

What is Reddit Answers?

Reddit Answers is a new AI-powered feature synthesizing information from existing posts and comments across Reddit's diverse communities. Think of it as having a smart assistant that can quickly scan through millions of Reddit conversations and provide concise, relevant answers to your questions.

A Redditor's Perspective - Me

As a long-time Reddit user, I've often dug deep into subreddits searching for authentic, unfiltered answers to my questions. What makes Reddit unique is its blend of anonymity and community wisdom - people feel free to share their genuine experiences and insights without the constraints of maintaining a professional image or brand presence.
While Reddit has always been a goldmine of information, finding specific answers often meant sifting through numerous threads, comments, and discussions. Sometimes, the most valuable insights would be buried deep within comment chains or scattered across multiple posts. This is where Reddit Answers becomes a game-changer.
The beauty of this AI-powered feature lies in its ability to aggregate and synthesize these scattered pieces of wisdom. However, as with all Reddit content, it's important to approach the information with a healthy dose of skepticism. The platform's anonymous nature, while enabling honest discussions, also means you should be mindful of potential misinformation from bots or trolls.
What makes Reddit Answers particularly exciting is how it maintains the platform's core strength - community-driven content - while making it more accessible through AI aggregation. Instead of replacing human interactions, it's enhancing our ability to tap into Reddit's collective knowledge base.

How Does It Work?

The process is remarkably straightforward:

Users navigate to reddit.com/answers or access it via the sidebar
They ask questions in plain English
The AI processes the query and generates responses based on real Reddit posts and comments
Each response includes citations linking back to sources

Here is an example screenshot,

Reddit Answers demonstrates AI-generated response about Medium article writing tips and best practices

Key Features and Limitations

Accessibility

Currently in beta testing in the United States
Available on reddit.com and the iOS app
English language only at launch

Usage Limits

Logged-out users: 10 questions per week
Logged-in users: 20 questions per day
Premium subscribers: 100 questions per day

Content Sources

Draws from public, non-NSFW communities
Excludes content from private, quarantined, and mature subreddits
Includes citation links to original posts and comments

The Technology Behind It

Reddit Answers utilizes generative AI combined with in-house technology to:

Find relevant content across communities
Synthesize information from multiple sources
Generate coherent, contextual responses
Maintain source attribution through citations

Privacy and Data Considerations

Reddit maintains transparency about data usage, adhering to its standard privacy policy. While responses can be shared via URL, usernames aren't associated with shared questions, protecting user privacy.

Why This Matters

This development is significant for several reasons:

Community Knowledge Accessibility: It makes Reddit's vast knowledge base more accessible and digestible.
AI Integration: Shows how traditional platforms can meaningfully integrate AI capabilities.
User Experience: Offers a new way to interact with Reddit's content beyond traditional search.

Looking Forward

While still in beta, Reddit Answers represents an interesting use case for generative AI in social platforms. It maintains the platform's community-driven nature while adding AI-powered convenience for information discovery.

For developers and AI enthusiasts, this implementation demonstrates how generative AI can be used to enhance existing platforms without replacing their core value proposition. It's a practical example of AI augmenting rather than replacing human-generated content.

Limitations and Considerations

Users should note that:

The AI may not always provide perfect summaries
Information should be verified, especially for important decisions
The feature is designed to summarize existing content, not generate new information
Response quality may vary, and feedback mechanisms are in place for improvement

Final Thoughts

Reddit Answers represents a thoughtful implementation of generative AI in a social platform context. It preserves the platform's community-driven nature while making its vast knowledge base more accessible. As the feature evolves beyond its beta phase, it will be interesting to see how it shapes the future of community knowledge sharing.

Want to learn more about practical AI tools and developments?

Follow The ABCs of GenAI publication for updates on the latest in accessible AI technology.
The ABCs of GenAI

Socials

Youtube
facebook.com/ABCsofAI
x.com/GorgeousTech
NEW! Podcast - AI Agents and Coffee

Use Prompt Engineering to Level Up Your AI Assistant when analysing long texts (Claude2)

T Frazao — Mon, 02 Oct 2023 01:00:00 +0000

Research-backed techniques to optimize prompts and few-shot learning

AI assistants like chatbots and voice assistants are becoming more capable every day. However, even advanced models like Claude can benefit from careful prompt engineering to get the most out of their skills.

In this post, I’ll share prompt engineering techniques you can use to enhance your AI assistant’s performance on tasks like:

Recalling information from long documents
Answering ambiguous questions
Avoiding biased word choices
Getting more creative and varied responses

With the right prompts, you can level up your AI assistant to be more helpful, knowledgeable, and human-like. I’ll provide examples and code samples for prompt techniques like using a , providing contextual examples, and structuring instructions clearly.

In summary, while tested on MCQs (multiple choice questions), these prompt engineering best practices can serve as guidelines for optimizing many different types of prompts and outputs with an AI assistant. The core principles apply broadly. Let me know if any part of my explanation needs clarification!

Whether you’re an AI developer or a casual user of AI chatbots, you’ll learn tangible strategies to improve your prompts. I will make it simple, I promise.

When asking Claude a question that requires recalling information from a long document, what can you do?

Note: before reading — it is not strictly necessary to explicitly label sections like “[Few Shot Examples]”, “[Prompt]”, and “[Passage]” or any other […] in the prompt example templates provided. The formatting is primarily for humans to read!

-->Use a <scratchpad> to extract key quotes from the document that are relevant to answering the question. This improves recall performance.

Scratch? what? Okay, you can stop scratching your head. This means <scratchpad> acts like a memory aid for Claude, created strategically by the user to call out the most relevant bits of the long context for answering the specific question. This technique improved Claude’s performance in recalling information from long documents in the prompt engineering experiments.

The workflow would be:

User provides a long document context
User identifies key quotes from the document that are relevant to answering the upcoming question
User inserts those quotes into a <scratchpad> block in the prompt
User then asks their question at the end of the prompt
When Claude processes the prompt, it can refer back to the <scratchpad> content extracted by the user to better recall where in the document context the information to answer the question is located.

Prompt Example Template:

Below is a 10,000 word document about the history of machine learning.

[insert long document here]

<scratchpad> Extract key quotes from the document that provide details about when deep learning started gaining popularity:

“In 2012, a deep learning model developed by Geoff Hinton and his students won the annual ImageNet image classification competition by a wide margin.”

“Between 2012 and 2015, deep learning began to be used for many more applications beyond image recognition, including natural language processing, audio recognition, and recommender systems.”

“In 2016, AlphaGo defeated the world champion in the game Go using deep neural networks, representing a major milestone in AI.” </scratchpad>

Question: When did deep learning start to become popular for applications beyond image recognition?_

The <scratchpad> extracts three relevant quotes from the document that provide context to answer the question of when deep learning expanded beyond image classification. This focuses Claude’s attention on the key information.

-->Provide several examples of other questions and answers from different parts of the document, to give Claude more context on the style of question you are asking.

The workflow:

Start with a long document or passage as the context.
Manually identify 3–5 short excerpts from different parts of the document that contain key details.
Write the question and answer pairs based on each excerpt, demonstrating how to query details from the passage.
Insert these question/answer examples sequentially in the prompt before the actual question you want Claude to answer.
Ask the target question at the end that you want Claude to respond to based on the full document.
Claude processes the entire prompt. The examples focus its attention on looking for similar types of details in the passage.
Claude applies this priming when reaching the target question, and leverages the full context to locate the relevant detail needed for its answer.

Prompt example template:

Below is a transcript of a speech given by the CEO of Company XYZ:

[Insert long speech transcript here]

[Examples:] Q: What does the CEO say is the company’s top priority for the year? A: The CEO states that the top priority is expanding internationally and entering new markets.

Q: How much does the CEO project revenue will increase next quarter? A: The CEO projects a 15% revenue increase next quarter.

Q: What competitive advantage does the CEO highlight in the speech? A: The CEO highlights their investment in R&D and proprietary technologies as a key competitive advantage.

Question: How many new products does the CEO announce they will be launching this year?

This provides Claude with three examples extracted from elsewhere in the long document, demonstrating the style of subjective questions that require picking out key details from the passage. The examples prime Claude on what type of information is being sought, improving its ability to recall the relevant detail from the document to answer the actual question.

-->Put the question and any instructions (like using the ) at the end of the prompt, so they are closest to where Claude needs to generate the answer.

The key is to position the most critical information (question + instructions) last, as close as possible to where Claude will produce its answer. This optimization aims to maximize Claude’s attention and recall when it matters most.

The workflow:

Provide the full context passage that the question will be about.
Include any examples or other directions earlier in the prompt.
Identify a key quote from the passage that provides useful context for answering the question.
Add the key quote within a block.
Craft a specific question that requires deducing the answer from the full passage, not just the quote.
Place the question and direction together at the very end of the prompt.
Claude will process the full prompt, prime its attention with the <scratchpad>, and then need to recall details from the entire passage to answer the question at the end.
Claude generates its response based on the full context.

The prompt example template:

[Context:] [Long passage describing a historical event]

[Examples:]

Q: When did this event occur? A: The passage states the event occurred in 1852.

Q: Where did the event take place?
A: According to the passage, the event took place in Boston, Massachusetts.

<scratchpad> “The passage mentions that there were over 200 people in attendance at the historical event.” </scratchpad>

Question: Approximately how many people were present at the event being described?

-->If the document context is very long (95,000+ tokens ~ 475,000 words, ~950 A4 pages), performance tends to dip slightly for information at the very end. So consider positioning key info earlier in the document if possible.

No applicable workflow or prompt example. Just organize the information in the document context.

-->When generating questions with Claude via Few Shot Learning, use a template that encourages Claude to specify details like which specific passage a question is about.
This avoids ambiguous phrases like “this document” when stitching passages together.
Few shot examples are a technique used in prompt engineering where you provide an AI assistant with a few examples demonstrating the task you want it to complete. This “primes” the model before asking it to generate new original content.

Specifically for question generation:

A few-shot example would show 2–3 sample questions about a passage, with the correct answers provided.
These examples demonstrate to the AI assistant the style and format of questions expected.
The prompt then instructs the assistant to generate a new question in that same style about a new passage.
By studying the few shot examples first, the AI can mimic that question/answer pattern for its new output.
Include guidelines in the Few Shot Learning prompt to prevent Claude from making the correct answer obviously different or more detailed than the wrong answers.
In the prompt itself, you would simply provide the examples directly, without labeling them as “few shot examples”.

Prompt example template:

[Few Shot Examples]

Q: What year did the Civil War begin, according to the passage about the origins of the Civil War? A: The passage about the origins of the Civil War states that the war began in 1861.

Q: How long did the battle last, as described in the passage about the Battle of Gettysburg? A: The passage about the Battle of Gettysburg indicates the battle lasted 3 days.

[Prompt:] Please generate a multiple choice question about the details contained in the passage below about the Lincoln-Douglas debates. When referencing the passage, specify “the passage about the Lincoln-Douglas debates” rather than using vague phrases like “this passage”. Provide 3 incorrect answers and 1 correct answer to the question.

[Passage:] [text about Lincoln-Douglas debates]

Include guidelines in the Few Shot Learning prompt to prevent Claude from making the correct answer obviously different or more detailed than the wrong answers

When using few-shot learning to have Claude generate multiple-choice questions, we want the answers to be fair and balanced. We don’t want Claude to make the right answer obviously different from the wrong one.

For example, we don’t want:

The right answer to be 5 sentences long and very detailed, while the wrong choices are only 1 sentence.
The right answer to sound more plausible or contain giveaway phrases like “this is correct”.

Instead, we want:

All answer choices to be similar length, ideally 1–2 sentences.
The right answer to seem equally possible as the wrong ones based on the passage.
No giveaway clues that reveal the answer.
To do this, we can provide examples in the few-shot prompt showing balanced answers. And give Claude instructions like:
Make all answers 1–2 sentences long.
Don’t include clues that give away the right answer.
All answers should seem possible based on the passage.
This way, picking the right answer requires deeper understanding, not just spotting surface clues.

TLDR:

Prompts are very important for AI assistants like Claude. Good prompts help Claude be smarter.
There are special techniques you can use in prompts to make Claude better at remembering and answering questions.
Some examples of good techniques are:
Using a to remind Claude of important details
Giving Claude examples of other questions and answers
Putting the question at the end of the prompt
Moving key info earlier in long passages
Asking Claude for balanced multiple-choice answers
Prompt engineering takes practice but can really improve Claude’s skills!
With the right prompts, Claude can be an even more helpful and knowledgeable assistant.
In summary, prompt engineering lets us level up our AI like Claude. There are proven techniques to make prompts optimized for different goals. Writing effective prompts is a useful skill for getting the most out of AI.

Notes:

Original article:

FREE READ LINK: https://medium.com/the-abcs-of-ai/use-prompt-engineering-to-level-up-your-ai-assistant-claude-b57cac293dcc?sk=83c4c7d0b00549b963d5b48c623ba4cf

I'm on a quest to make AI less scary, one Medium post at a time. Join me on this totally uplifting journey by following my publication and subscribe to the weekly newsletter.

ABCs of AI
https://medium.com/the-abcs-of-ai

How did you like this post?
You can reach me at contact@abcsofai.info. I welcome your feedback and suggestions. And,

Follow me on the socials:

Linkedin: www.linkedin.com/in/tfmissgorgeoustech
Facebook: facebook/ABCsofAI
Youtube: @ABCs-of-AI
X: @GorgeousTech

Thank you.

Learn AI for Free in 2023: The Expert List of Resources, Guidance, and Announced AI Events (in progress)

T Frazao — Tue, 26 Sep 2023 16:50:36 +0000

Hey there! I wanted to share a little sneak peek at a blog post I’m cooking up with a bunch of excellent free resources for learning about artificial intelligence and machine learning.

Now, I know AI seems intimidating — all those complex algorithms and tech lingo. But have no fear! I’ve been digging around to find top-notch courses, tutorials, and tools perfect for beginners and not so total. I’m bookmarking all my faves to pass along to you.

Consider this your starter roadmap to launch into the world of AI for free! I’ll be adding to it as I discover more gems in my own AI learning adventures.

To kick things off, I highlighted some foundational programs from IBM, AI Planet, and CS50AI. Check out the links in bold and underlined when you get a chance! I’m still developing the full content, but wanted to give you this little teaser.

Make sure to check back for updates as I add more resources. And let me know about any favorite free learning tools you think I should feature! The more we share, the better we all get at AI.

Oh, and I’m including a section on major AI events and conferences coming up. These are fantastic for learning about new research and innovations. More on that later!

For now, hopefully, this gives you a boost to fuel your AI education. Let the learning begin, right?

Happening Now — Don’t Miss Out!

AI Planet’s LLM Bootcamp
It is currently underway — AI Planet recently launched an exciting free Large Language Model (LLM) Bootcamp from August 12th — September 24th 2023 (but you can do the projects till October 24th 2023) to help the community stay on top of the latest AI advances.

The bootcamp will explore cutting-edge transformers, open-source LLMs, and frameworks like Langchain and LlamaIndex through live expert sessions and hands-on projects.

Registration is FREE — just sign up at https://aiplanet.com/bootcamps/llm-bootcamp!

Even if it has already begun you can still sign up — Ideal for NLP enthusiasts, data scientists, and developers, the LLM bootcamp will cover topics like attention mechanisms, transforming for NLP tasks, and applying Langchain and LlamaIndex. Don’t miss this free opportunity to dive into the fascinating world of large language models and get hands-on practice with industry-leading AI.

Events: AI Conferences Happening Soon

September 27–28, 2023 — Meta Connect 2023 — Online and in VR Don’t miss the Meta Connect 2023 lightning talks and sessions highlighting Meta’s latest AI innovations. On Day 2, there will be talks diving into the Llama ecosystem for natural language processing, building next-gen AI experiences with Llama, taking Llama to an enterprise level, and new developments in Meta Spark for mobile. Attend these sessions to learn about Meta’s progress in contextual AI, large language models like Llama, and applying AI advancements to create new products and experiences.

There are two ways to watch Meta Connect 2023:

Online: You can watch the Meta Connect keynote and other sessions on Meta’s website, Facebook page, YouTube channel, and Twitch channel. The keynote starts at 10:00 AM PT / 1:00 PM ET on September 27, 2023.
In VR: You can also watch Meta Connect 2023 in VR using the Meta Horizon Worlds app on your Meta Quest 2 headset. This will give you a more immersive experience, and you’ll be able to interact with other attendees and the event space.
To watch in VR, update your headset, download the Meta Horizon Worlds app, and select the Meta Connect 2023 event.

Registration for the virtual or VR event is free at metaconnect.com.

September 28, 2023 - Datacamp Radar Conference— Online DataCamp is hosting a special edition of their RADAR conference. It’s totally free to attend! They’ve gathered data and AI experts to chat about building critical data skills for the digital age we live in. You know, stuff like machine learning, data visualization, Python programming, and more.

So if you’re interested in future-proofing your career and learning from some smart people in the industry, be sure to register for RADAR.

You can register for the event at Datacamp Event

October 8–10, 2023- AI Engineer Summit — In Person and Online I’m thrilled to share an exciting upcoming event for anyone fascinated by artificial intelligence.

Billed as the “Solvay Conference for AI Engineers,” this premier gathering aims to unite leading minds advancing the AI field, just as the 1911 Solvay Conference brought together pioneering physicists like Einstein.

The Summit agenda features cutting-edge research presentations, lively discussions, and plenty of networking among top AI talent. Attending this influential conference provides an incredible opportunity to gain insider perspectives on new innovations and the future of artificial intelligence directly from those shaping it.

While in-person tickets are limited, you can also participate virtually to live stream sessions as they happen.

Check the link here.

November 6, 2023 - OpenAI’s first DevDay conference — In Person and Online

Mark your calendar for OpenAI’s first DevDay conference on November 6, 2023 in San Francisco — the premier event for AI developers this year.

Sign up using the link below to receive email notifications when in-person registration opens in the coming weeks. Spots are limited, so be ready to secure your place at this groundbreaking gathering.

Can’t attend? The keynote will be livestreamed worldwide so you can tune in for major announcements.

Learn on Your Schedule: Self-Paced AI Courses

- CS50 — AI
I really love CS50. In three words, is FREE, very well-designed, and practical.

This course has had an update on LLMs on August 2023. The tail end of the course has been refreshed to incorporate discussion of large language models like ChatGPT. Aside from changes to the code submission branches, which were updated across all projects, only Lecture 6 and Project 6 underwent significant revisions. The course seems designed to allow a smooth transition for those who started before the 14 August 2023 update. Both versions of Project 6 will be accepted for credit through September 2023.

Now, you have the option to try this course. It already started before 2023. The course’s lectures were filmed with Brian using Python 3.8. You should use Python 3.8, 3.9, or 3.10 to work on the course’s projects. You should not use Python 3.11 or newer, due to its incompatibility with some of the modules used in later projects.

The course provides an introduction to the fundamental concepts and algorithms that power modern AI, including search algorithms, classification, optimization, machine learning, and natural language processing.

Check the link - CS50’s Introduction to Artificial Intelligence with Python

Students gain hands-on experience by implementing Python programs that apply key AI principles. Through coding projects and assignments, they acquire practical skills with AI libraries and frameworks while exploring the theoretical foundations of intelligent systems. The goal is to equip learners with core AI knowledge to enable them to build their own intelligent applications.

What about the deadline which is 31 December 2023? Do not worry, just do it. On 1 January 2024, the course will be updated to a 2024 version. Your progress from 2023 will be carried forward to the 2024 version. After 31 December 2023, you may only submit the 2024 version’s problem sets and will need to satisfy whatever requirements are in place for CS50x 2024.

Overall, this looks amazing way to get an introduction to key AI algorithms and techniques, with lots of practical applications through projects. The course strikes a nice balance between theory and hands-on coding. I’m excited to check it out as I continue expanding my AI skills.

-IBM — Jumpstart Your AI Learning with IBM
IBM offers one of the most comprehensive, utterly free AI education programs out there. The best part is it takes you from theory to practice with a range of courses for different levels. As a beginner, you can start with their introductory courses to get a solid base of knowledge on key concepts, all for free. Once you have the basics down, you can dive into hands-on guided projects to gain practical experience with popular techniques like generative AI and reinforcement learning, without having to pay a cent. IBM has top experts design these free courses and projects to ensure you are learning industry-relevant skills on the cutting edge. Having real coding practice on your resume will give you a significant leg up. Whether you just want to learn for fun or launch a new AI career, IBM provides the full curriculum you need to become both an AI expert and practitioner, entirely free of charge.

Now that you know IBM offers a comprehensive free AI education program, let’s dive into what’s included. Below you’ll find the complete list of courses and hands-on projects comprising their Fundamentals of AI Learning Path. I’ll summarize each course and project so you can get a sense of the skills you’ll develop and the topics covered at each stage, from beginner foundations to intermediate practice. Whether you’re new to artificial intelligence or looking to expand your skills, this curriculum provides the knowledge and hands-on experience to succeed — 100% free!

The IBM AI Learning Path contains the following 3 introductory courses:

Introducing AI: Covers AI concepts, real-world impact, and generative AI (AI that can generate new content like images, audio, and text) applications.
AI Concepts: Explains core AI terminology and techniques like machine learning and neural networks.
AI Ethics: Discusses ethical considerations around bias, transparency, regulations, and trust.

To get hands-on intermediate experience, IBM recommends 3 guided projects focused on generative AI (AI that can generate new, original content like images, audio, and text):

The Art of Prompt Engineering: Learn to optimize prompts and datasets for AI writing assistants.
Create Your Own Chatbot Website: Build a chatbot with open-source large language models that can generate conversational text.
Creating Anime Characters using DCGANs: Generate new anime images with deep convolutional GANs (generative adversarial networks).
Natural Language Processing (NLP) with Hugging Face Transformers: Apply NLP techniques like text classification and summarization with Hugging Face.
Build an Image Style Transfer Tool using CycleGANs (generative adversarial networks): Implement style transfer between images with CycleGANs.
Create a Voice Assistant with OpenAI’s GPT-3 (Generative Pre-trained Transformer 3) and IBM Watson: Build a voice assistant using GPT-3 and IBM Watson APIs.
Create AI powered apps with open source LangChain (foundation model for conversational AI): Build AI apps with the LangChain library for chatbots and generation.

- Text Sentiment Analysis using Caikit and Hugging Face transformers:
Classify text sentiment with Caikit and Hugging Face transformers.

There are also projects recommended for reinforcement learning (training AI agents by having them interact with an environment) and time series analysis (modeling and forecasting data over time).

Reinforcement Learning Projects:

Monte Carlo Reinforcement Learning for Simple Games: Implement Monte Carlo methods to play simple games.
Win Blackjack with Reinforcement Learning: Build an agent to play blackjack through reinforcement learning.
Playing TicTacToe with Reinforcement Learning: Create a TicTacToe AI agent with reinforcement learning.
Unleashing the Power of Reinforcement Learning for Trading: Apply reinforcement learning techniques for algorithmic trading strategies.

Time Series Projects:

Machine Learning for Sequential Data
: Apply ML models like RNNs (recurrent neural networks, a type of neural network well-suited for sequential data) to sequential time series data.
Time Series Forecasting With Prophet: Use Facebook Prophet for time series forecasting.
Create a Cryptocurrency Trading Algorithm
: Build a crypto trading bot with time series models.
Forecasting Crypto Portfolios Like a Quant: Use quant strategies for crypto portfolio predictions.

And that wraps up our roundup of quality free resources to launch your AI education! I hope this gives you plenty of options to start mastering artificial intelligence and machine learning fundamentals. Make sure to bookmark your favorites so you can take advantage of these courses, tutorials, and tools.

Check Out My Other Articles on my Medium's Publication (https://medium.com/the-abcs-of-ai/) and follow me on the socials @GorgeousTech

Good News Ahead: An ABCs of AI Publication Newsletter -1º edition - (Preview)

Zoom AI Companion: A New AI Assistant for Boosting Productivity

A Personal Encounter with the MetaMentor Plugin from AxonAI

Gandalf’s Challenge: Mastering Prompt Engineering for AI Success

Boost Your Skills with AI-Assisted Data Analytics

AI in Business: How McKinsey Harnesses the Power of ChatGPT

The Future of Chatbots: OpenAI’s ChatGPT Roadmap Explored

The Evolution of Artificial Intelligence: A Deep Dive into the History of AI

Overcoming Anxiety in the AI Era: Harnessing the Power of AI Positively

and the list continues...

Thank you for reading this post and enjoy learning with me.