🌟 Key Highlights from the Announcement

Ensuring Security on a Global Scale 🛡️:

The regulations will institute comprehensive testing for advanced AI products, ensuring they cannot be repurposed for the production of biological or nuclear weapons. This move underscores a steadfast commitment to global security.

Safeguarding Against “Deep Fakes” 🎭:

Recommendations for watermarking AI-generated content will be introduced to combat the rising threat of “deep fakes” and disinformation campaigns. This measure is particularly pertinent as we approach the 2024 presidential campaign.

A Call for Global Collaboration 🌍:

Recognizing the global nature of AI development, the U.S. is proactively encouraging allies and adversaries alike to develop parallel regulations. This reflects the imperative of international cooperation in regulating AI technologies.

Influence on Technology Sector 💡:

The regulations will pioneer standards for safety, security, and consumer protections within the technology sector. By leveraging the government’s influence, compliance with these standards will be incentivized among technology companies.

Advancing Clear Safety Standards 📋:

The Department of Health and Human Services, in collaboration with other agencies, will set forth unequivocal safety standards for AI usage. Additionally, procurement processes for AI tools will be streamlined, ensuring accessibility and responsible adoption.

Evaluating the Impact on Labor Market 💼:

The Department of Labor and the National Economic Council will undertake a comprehensive study to assess AI’s effect on the job market. This study may lead to further regulations, ensuring the benefits of AI are harmoniously balanced with workforce considerations.

Mitigating Algorithmic Bias 🚫:

Agencies will provide clear guidance to prevent discrimination resulting from algorithms used in AI tools. This measure will have positive implications for landlords, government contractors, and federal benefits programs.

🔍 Further Considerations

While these regulations represent a significant stride, it’s important to note that the White House acknowledges the necessity for privacy legislation to fully ensure robust data protection. The tech industry, including Microsoft, OpenAI, Google, and Meta, stands united in supporting these regulations, albeit with differing opinions on the level of government oversight.

🔮 The Road Ahead

Looking ahead, this landmark development sets the stage for a new era of responsible innovation. President Biden’s commitment to finding the equilibrium between innovation and regulation in the AI sector is emblematic of a forward-thinking approach. These regulations not only aim to harness the benefits of AI but also to mitigate potential risks, thereby paving the way for responsible AI development and usage.

Reach out to me

Use the links below to check out my other content or just to say hi.

Check out my other articles on Medium: https://medium.com/@namyagupta2001

Reach out to me on LinkedIn. Let’s connect: https://www.linkedin.com/in/namya-aankur-gupta/

Be sure to check out AI Banter’s youtube video and Subscribe 🚀 : https://www.youtube.com/watch?v=u76m8lWvc3s&t=607s

Instagram: https://www.instagram.com/ai.banter/?hl=en

Twitter: https://twitter.com/AankurNamya

Bloomberg has ushered in a new era with the unveiling of BloombergGPT, an exceptional large language model meticulously tailored to cater to the intricacies of the financial realm. Let’s dive into the phenomenon that is BloombergGPT and its game-changing implications for the finance industry.

Deciphering the Finance Lexicon

Navigating the intricate language and data-driven essence of the financial world demands an AI solution that understands its intricacies. While general large language models have laid the groundwork, BloombergGPT raises the bar by undergoing intensive training on an extensive corpus of financial data. This unique training equips it to excel in diverse natural language processing tasks specific to the finance domain.

A Legacy of AI Pioneering

Bloomberg’s decade-long leadership in AI and machine learning comes to fruition with BloombergGPT. This model ingeniously merges domain-specific financial data with general-purpose datasets, giving rise to a powerhouse that not only outperforms peers in financial benchmarks but also stands strong in broader language benchmarks.

Power of Quality Data

The foundation of any great AI model is the data it’s nurtured with. Drawing from an extensive archive of four decades of financial data, Bloomberg curated an exhaustive dataset comprising a staggering 363 billion tokens. This in-house data was further enriched with a 345 billion token public dataset scraped from diverse sources like YouTube and Wikipedia. This meticulous curation resulted in a colossal training corpus exceeding 700 billion tokens.

Bloomberg uses Natural Language Processing to transform how users find financial data, Credits: Fortune

BloombergGPT’s Multifaceted Abilities

BloombergGPT shines brilliantly across an array of tasks:

1. Mastering Bloomberg Query Language (BQL): Simplifying financial analysis, it translates natural language queries into valid Bloomberg Query Language, enhancing data interactions.
2. News Headline Wizardry: Assisting journalists and news platforms, it suggests compelling and contextually relevant headlines that captivate readers.
3. Financial Q&A Maestro: Proficiently addressing financial queries, BloombergGPT offers precise and insightful answers, showcasing its mastery of the domain.

AI’s Unstoppable Surge in Finance 📈

In a landscape dominated by data’s sheer magnitude and complexity, AI’s indomitable role in the financial sector is undeniable. Bloomberg’s launch of BloombergGPT on March 30, 2023, is a testament to this evolution, propelling AI into the heart of finance.

AI’s Multifaceted Role: As financial institutions grapple with data’s exponential growth, AI offers transformative solutions across key domains:

Revolutionizing Chatbots: AI-powered chatbots, empowered by Natural Language Processing, redefine customer engagement, offering round-the-clock interactions and enhanced online services.

Fortifying Fraud Defence: AI augments fraud detection by unveiling intricate patterns, anomalies, and suspicious relationships within data, bolstering existing prevention measures.

Empowering Credit Risk Management: Amidst stringent regulations, AI emerges as the driving force behind robust credit risk management models.

Predictive Precision: AI and Machine Learning usher in accurate revenue forecasts, stock price predictions, risk assessment, and case management.

Enhancing Customer Relations: AI’s personalized touch translates into 24/7 services, featuring biometric authentication and voice-command capabilities.

Revolutionary Architecture and Beyond 🦾

BloombergGPT’s architecture, a decoder-based causal language model, boasts an astounding 70 layers of transformer decoder blocks. Its ability to seamlessly convert natural language into Bloomberg Query Language streamlines financial analysis, while its knack for generating news headlines demonstrates its adaptability.

Reshaping Finance with BloombergGPT 👩🏻‍💻

BloombergGPT’s debut marks a monumental leap in the fusion of finance and AI. Armed with 50 billion parameters and unparalleled domain-specific finesse, it emerges as a trailblazer in financial NLP tasks. As research and training continue, BloombergGPT’s potential to reshape the finance landscape remains boundless.

Watch My YouTube Video to Dive Deeper into BloombergGPT’s Impact on Finance!

Stay tuned for more transformative developments at the convergence of technology and finance!

Reach out to me

Use the links below to check out my other content or just to say hi.

Check out my other articles on Medium: https://medium.com/@namyagupta2001

Reach out to me on LinkedIn. Let’s connect: https://www.linkedin.com/in/namya-aankur-gupta/

Be sure to check out AI Banter’s youtube video and Subscribe 🚀 : https://www.youtube.com/watch?v=u76m8lWvc3s&t=607s

Instagram: https://www.instagram.com/ai.banter/?hl=en

Twitter: https://twitter.com/AankurNamya

The reason for this big growth is that AI is being used in more and more ways in different industries. It’s changing how we do things in healthcare, transportation, and more. AI is making a big difference in how we live and work, and it’s even changing what we thought was possible. The demand for the computer chips that power AI is going up a lot, and companies like NVIDIA are making a lot of money from it. Recent product and service announcements from Nvidia CEO Jensen Huang demonstrate the company’s dedication to growing its AI footprint.

GEForce RTX GPU

Earlier image rendering used to take hours until NVIDIA came up with a technology to develop the image rendering process to be accomplished in minutes. Now, with NVIDIA’s GeForce GPU, the image rendering process has reduced to 15 seconds approximately. It works with the help of AI and algorithms to show proper shadows as well as 3D rendering of GPU. It has made a very significant impact by building powerful GPU’s that have helped the data scientists and engineers to train their machine learning models by boosting the performances and thus, making them rely on NVIDIA.

NVIDIA ACE (Avatar Cloud Engine)

With the recent unveiling of NVIDIA’s ACE (Avatar Cloud Engine), the horizon of gaming appears to be expanding exponentially. It has the capabilities of pioneering AI engine, demonstrating its prowess in speech recognition, text-to-speech, and natural language understanding through a monumental language model for gaming. It will animate the gestures by the voice modulation of the speaker. They have partnered with ConvAI to build the powerful framework for beautiful 3D images and videos. It involves generative AI with which the non playing characters (NPCs) can interact and communicate with the characters in the game.

DGX GH200 AI Supercomputer

The introduction of AI supercomputer DGX GH200 marks its continuous efforts to cater to the needs of datacenter operations. The DGX GH200 is not just a fancy piece of rack hardware; it’s a comprehensive solution that combines high-performance computing (HPC) with AI. It’s designed to handle the most complex AI workloads, offering a level of performance that is truly unparalleled.

The DGX GH200 has all the raw power developers could want; it is much more than just a fancy super-computer. It’s about harnessing that power to drive AI forward. Undoubtedly, the software stack that comes bundled with the DGX GH200 is one of its standout features. This comprehensive solution comprises several optimized SDKs, libraries, and tools designed to fully harness the hardware’s capabilities, ensuring efficient application scaling and improved performance.

A wider democratic growth of AI technology may result if more large businesses adopt this technology and start to develop and implement cutting-edge AI solutions. Hopefully, new developments will filter down into more affordable solutions, opening up AI to smaller businesses. The ability for smaller organisations to use computational capacity comparable to the DGX GH200 on a pay-per-use basis is expanding thanks to cloud-based connectivity. The DGX GH200 may have a high initial cost, but its long-term impact on the industry may help level the playing field for companies of all sizes. Jensen gave an example of Synthesia AI that converts the Text to Actor narration. You can click on the link to see the demo.

✂️ DGX GH200

NVIDIA Maxine

NVIDIA Maxine has been developed specifically to address these challenges through the use of state-of-the-art AI models that greatly improve the clarity of video conferencing calls. NVIDIA Maxine is a suite of pretrained AI models built to improve video conferencing experience. NVIDIA is accelerating its effort to provide cloud-native microservices to enable disaggregating computing in the cloud, for “scale out” beyond a single GPU and improving resource management. NVIDIA Maxine microservices can be integrated with your existing software and deployed in Kubernetes clusters with GPUs in the cloud. This also simplifies deployment for cloud infrastructure and provides the ability for companies to leverage Maxine in private or public cloud infrastructures.

Three types of microservices are provided:

• Audio Effects microservice: Includes background noise removal, room echo cancellation, acoustic echo cancellation and audio super resolution
• Video Effects microservice: Includes virtual background and eye contact
• Live Portrait microservice: For animating a picture from your webcam feed input

NVIDIA and WPP

NVIDIA and WPP announced that they are creating a content engine that uses AI and NVIDIA Omniverse to help creative teams produce high-quality commercial content at scale, more quickly, and with more brand consistency.

The new engine links an ecosystem of tools for 3D design, manufacturing, and the creative supply chain, including those from Adobe and Getty Images, enabling WPP’s designers and artists to combine the development of 3D content with generative AI. This makes it possible for their businesses to engage with customers in highly personalised ways while maintaining the integrity of their brand, its products, and its trademarks.

✂️ Generative AI advertisements

NVIDIA keeps pushing the envelope of AI innovation with the release of the DGX GH200 supercomputer platform, the partnership with WPP for the production of advertisements, and Nvidia ACE for Games 🚀 Nvidia strengthens its position as a pioneer in the field and advances the AI industry by offering cutting-edge technologies and tackling significant issues in AI development.

If this newsletter intrigued you about AI, share it with your network too! Happy reading!

Reach out to me

Use the links below to check out my other content or just to say hi.

Check out my other articles on Medium: https://medium.com/@namyagupta2001

Reach out to me on LinkedIn. Let’s connect: https://www.linkedin.com/in/namya-aankur-gupta/

My Twitter: https://twitter.com/AankurNamya

Be sure to check out AI Banter’s youtube video and Subscribe 🚀 : https://www.youtube.com/watch?v=u76m8lWvc3s&t=607s

Instagram: https://www.instagram.com/ai.banter/?hl=en

Expanding AI Horizons with Meta’s LlaMa 2 👩‍💻

• On July 19, Meta, founded by Mark Zuckerberg, declared that it would collaborate with Microsoft to launch the next-generation open-source language model ‘LLaMa 2’.
• LLaMa 2 is a large language model (LLM) that can be used to create chatbots such as ChatGPT.
• Meta is making LLaMa 2 open-source, so, its content will be available for people to access for free and use and tweak to create their own AI chatbots for other purposes.

Microsoft Got the Move 🏎️

• Microsoft has invested over $10 billion in OpenAI, but its partnership with Meta indicates that it is broadening its AI bets beyond ChatGPT.
• Open AI, co-founded by Elon Musk and investor Sam Altman and backed by $1 billion in funding from Microsoft Corp., is expecting its business to surge.
• The research organisation expects $200 million in revenue next year and $1 billion by 2024.

AI-vertising for Meta 📺

• This can prove to be a new source of revenue for Meta alongside their primary source — Advertising and can also help them tap into new markets.
• By open-sourcing LLaMa 2, Meta gains a tactical advantage by preempting potential customers who might have turned to competitors like Google or Microsoft.

Apple GPT 🧑‍🔬

According to reports, Apple is creating the “Apple GPT” AI chatbot. These models are produced by Ajax framework, which intriguingly functions on Google Cloud.

Only a small number of Apple employees have access to the Generative AI technology, which is presently in testing. This report is even more surprising given how secretive the tech giant has been about its research in generative AI.

The tech industry will undoubtedly be affected by the iPhone maker’s entry into the field of generative AI, which was previously dominated by OpenAI and Microsoft. You can expect significant upgrades to Siri.

Google’s Bard Gets Eyes 👀

Bard has been updated with image recognition capabilities.

This means that Bard can now understand and respond to images, opening up a whole new world of possibilities.

Google’s New AI News Tool- Genesis 📰

Google is developing an AI tool called “Genesis” that could revolutionize the news industry. It is capable of writing news articles & potentially replacing 43% of journalists.

The tool, known internally by the working title Genesis, can take in information — details of current events, for example — and generate news content, the people said, speaking on the condition of anonymity to discuss the product.

One of the three people familiar with the product said that Google believed it could serve as a kind of personal assistant for journalists, automating some tasks to free up time for others, and that the company saw it as responsible technology that could help steer the publishing industry away from the pitfalls of generative A.I.

🛡ChatGPT 4 vs LLaMa 2 vs Bard LLM ⚔️

• Meta is trying to weaken the competitive edge of OpenAI, Google etc. Interestingly, Microsoft is the biggest investor in OpenAI, yet it is backing Llama 2.
• ChatGPT took off like a rocket in December’22 and grew 132% in January’23 and then had a slow growth in April-May. Meanwhile, Google’s Bard attracted 142.6 million visitors in May, up from 49.7 million in April.
• Both Bard and the Bing chatbot (which incorporates OpenAI’s latest GPT-4 model, are free alternatives that have the advantage of access to more up-to-date information than ChatGPT because they are built on top of search engines.
• Llama 2 is Open Source as well as Free for commercial use 🚀 This means that the companies that had concerns over sending their data to OpenAI will now benefit from Llama 2.

AI Odyssey Continues 🔮

• The release of consumer-focused AI tools such as ChatGPT and Bard is set to fuel a decade-long boom that grows the market for generative AI to an estimated $1.3 trillion in revenue by 2032 from $40 billion last year.
• The sector could expand at a rate of 42% over ten years — driven first by the demand for infrastructure necessary to train AI systems and then the ensuing devices that use Al models, advertising, and other services.

Amazon.com Inc.’s cloud division, Google parent Alphabet Inc., Nvidia Corporation, Microsoft Corporation and Apple, that have invested billions of dollars in AI, are likely to be among the biggest winners from the Al boom.

Thanks for Reading!

Reach out to me

Use the links below to check out my other content or just to say hi.

Check out my other articles on Medium: https://medium.com/@namyagupta2001

Reach out to me on LinkedIn. Let’s connect: https://www.linkedin.com/in/namya-aankur-gupta/

My Twitter: https://twitter.com/AankurNamya

Be sure to check out AI Banter: https://www.youtube.com/@AI-Banter

Unveiling the Art of Language Modelling

Language models are becoming robust in the field of natural language processing (NLP), capable of comprehending and producing writing that is human-like. One fundamental practise stands out among the many strategies that support their success: Prompt Engineering.

This intriguing method involves developing prompts — effective instructions or queries — to direct language models towards desired outcomes.

Let’s explore prompt engineering and learn about its tremendous effects in the NLP field

Use Cases

Text Generation: Language models are able to produce text that adheres to a variety of styles, tones, and topics by adjusting the prompts. This talent is very beneficial for personalised chatbots, content creation, and creative writing.
Answering questions: Prompts may be created to draw forth pertinent data from a particular situation. This method improves the accuracy and context-appropriateness of the replies provided by chatbots, information retrieval systems, and virtual assistants.
Sentiment Analysis: Language models may be adjusted to analyse sentiment in text data by including prompts that specifically ask for sentiment-based replies. This supports sentiment analysis, brand monitoring, and perception research.
Machine Translation: Prompts can be used to direct language models towards creating accurate and contextually relevant translations when using machine translation. This makes machine translation systems better and makes multilingual communication easier.

Successful Prompt

To help with content development, a simple prompt template may include words, phrases, and questions. The target audience, tone of voice, and style, as well as a strong call to action, are components of a successful prompt template, on the other hand.

Prompt Engineer

Role of a Prompt Engineer

A prompt engineer is required to:

• Design of Prompts: A prompt engineer creates prompts that provide language models clear instructions. This entails comprehending the precise specifications of the NLP task at hand and developing prompts that provide the necessary results. Examples, inquiries, or detailed directions that are suited to the task may be included in the prompts.
• Task-Specific Optimisation: To enhance the performance of language models for certain NLP tasks, prompts are optimised by the prompt engineer. This involves repeatedly experimenting with and analysing the prompts in order to improve and fine-tune them. They seek to improve accuracy, coherence, and contextual relevance by assessing the model’s output and tweaking the prompts.
• Analysis of Data: Engineers who work quickly analyse and preprocess data to find trends, insights, and potential biases. They examine the training data, perform any necessary preprocessing, and look for any particular patterns or features that might be added to the prompts to enhance model performance. Additionally, they guarantee that the data employed for prompt engineering is diverse and representative.

The salaries of Prompt Engineers range from $90,000 — $320,000 in the United States of America

Who can be a Prompt Engineer?

You can be a prompt engineer if you have the following skills broadly:

• Good verbal and written communication: Effective communication is crucial when working as a prompt engineer. You should be able to clearly communicate ideas, collaborate with team members, and convey complex technical concepts to stakeholders with varying levels of expertise.
• AI technical knowledge: You have a solid understanding of machine learning principles, including supervised and unsupervised learning, as well as deep learning techniques like neural networks. Knowledge of training algorithms, optimization methods, and hyperparameter tuning is beneficial.
• Data Analysis Experience: Prompt engineers work with data to identify patterns, insights, and potential biases. They analyze the training data, preprocess it as required, and identify any specific patterns or features that can be incorporated into the prompts to improve model performance. They also ensure that the data used for prompt engineering is representative and diverse.

My experience with Prompt engineering

The reliance on many individuals or jobs has been completely decreased, and I can now perform at least 4 to 7 responsibilities on my own using these amazing AI capabilities with the help of Prompt Engineering.

Here is what I have learnt to give a good prompt:

Give OpenAI’s ChatGPT a context, then add the work you want it to do. Include step-by-step instructions and offer the data it needs to produce the desired result (content, code, an image, analysis, etc.). Give the AI a position that is appropriate for the results you expect, such as content producer, artist, proofreader, analyst, developer, author, blogger, strategist, or marketer. The prompt should be improved till it produces the desired result. Give examples, influencers, or well-known figures in those fields so that you can produce superior content or production.

For image creation provide the image reference, text prompt, and parameters like lighting, 3D, photo realistic, cinematic, and negative prompt what it should not do. This gives a clear guidance for the tool to generate the desired output.

For creating AI based videos — create an image, add the script, and use the AI video creating tools to generate the videos. You can edit, remove unwanted content, delete the long pause etc., make the videos better and engaging.I have been having a great time with these tools and experimenting every day. I could create images, edit videos, content, art, etc.

This industry is expanding quickly, and newer and better products are being introduced daily. Although it is challenging for everyone to be aware of everything, failing to be informed and use these tools is a crime. Getting your hands on these tools and improving the speed and quality of your work is the only way to keep your roles. Boost your productivity and worth!

Thanks for Reading!

Reach out to me

Use the links below to check out my other content or just to say hi.

Check out my other articles on Medium: https://medium.com/@namyagupta2001

Reach out to me on LinkedIn. Let’s connect: https://www.linkedin.com/in/namya-aankur-gupta/

My Twitter: https://twitter.com/AankurNamya

Be sure to check out AI Banter (content coming soon 🔥): https://www.youtube.com/@AI-Banter

ChatGPT is one of the most trending topics right now. It has broken the record for most sign ups over 1 Million users done even more than Netflix, Youtube and Instagram in just 5 days. No longer we will be googling the answers to problems, it gives breakdown of them in fine detail with explanations on how it works. Example, please summarise this text for me (and you enter the text) , it has you covered!

It is built by OpenAI, they are shaping some really interesting technologies. They have built CodeX, engine behind GitHub Copilot. You might have seen people putting prompts and getting the outputs from an AI generated system. Similarly, one more technology is DALLE-2, an image generator. Here we get different types of visualisations.

ChatGPT, is therefore a way of prompting AI to give the solution in a human like manner. This can be scary for organisations like StackOverflow, which have themselves associated with human related answers. They have actually have banned ChatGPT out from their system. If people simply copy paste answer from GPT without any references, it will break the trust. As of now, it does not have data after 2021, so it can’t tell you about the FIFA World Cup, for example.

You can use chatGPT by going to this link and signup if you have not already in order to try it.

Introducing ChatGPT

Natural Language Processing and Generative Pre trained Transformer

ChatGPT is a Natural Language Processing model known as the Generative Pre trained Transformer(GPT). Here, you will also get to know about Supervised learning and Reinforcement learning.

Natural Language Processing is the combination of Linguistics and Computer Science. Autocorrect, autosuggest, plagiarism checks make use of NLP. It is subfield of Artificial Intelligence that understands and generates human language. ChatGPT understands what we are saying and generates the solutions. Computers in general don’t know any language besides numbers and vectors. The GPT Encoder finds its own structure and patterns in the data it gets provided.

Every word gets converted to vectors of numbers called speech tagging. The better it understands, the better it can use this knowledge to generate its own text similar to human language. This is the way GPT works. Generative means the model can generate something. Pre trained means that model has been trained initially on vast amounts of data. The data has been cutoff in September 2021. The Transformer made this whole technological breakthrough in the first place! It uses the encoder to give the numerical representation of the input sentence. It then captures the meaning of the sentence from sequence to sequence. The breakthrough of the Transformer is that its sequence processing is highly parallelizable.

It calculates how much attention each vector should be given for a given sentence. It allows the model to effectively process input sequence of any length in a parallel and efficient manner. It vastly outperforms RNNs and CNNs that are slow for this case.Before the model gets fine-tuned, it is first trained by predicting removed words on unsupervised data.

Supervised Learning and Reinforcement Learning

Initial model was trained through supervised fine tuning. A lot of prompts were created and humans described the desired input. It helps to basically understand what’s going on and getting trained.

Reward model was also used, for example, a prompt gives 4 options to an answer, then ranks those options with labels from best to worst. This data is then fetched to the model. Finally, it is about optimising a policy against reward model using reinforcement learning.

Model tries to teach itself how to get better by asking the questions how can I do it right, what did I do wrong. Thus, gives off the best results.

It will usher in a new paradigm of AI powered tools that will do incredibly impressive things over the coming years!

If you liked the article and you want to know more about cool tech stuff, or you just want to ask me something, you can: follow me on LinkedIn, where I publish all my stories, subscribe over here on Medium.

Unleashing the Power of ChatGPT — A Revolutionary AI Powered Chatbot was originally published in BLOCK6 on Medium, where people are continuing the conversation by highlighting and responding to this story.

NFTs and digital collectibles are growing popular as the web3 space continues to make significant advancements in the blockchain arena. The enormous popularity of NFTs like Cryptokitties and Bored APE pushed investors to purchase ERC721-compatible digital collectibles.

In this article, we will guide you through the procedure of creating your first ERC-721 (NFT) contract on Solidity with Open Zeppelin.

Developing a Smart Contract with Solidity and deploying it on the blockchain might sound daunting at first: solidity, security, gas optimisation, developer environment, and gas fees, are only some of the things you’ll need to go through to host your code on the blockchain.

Tools like the OpenZeppelin Wizard that offers developers click and write functionalities to create composable and secure smart contracts in no time, used with Web3 developer tools like Alchemy, make the experience of writing a deploying code on the blockchain easy, fast, and reliable like never before.

In this tutorial, you’re going to learn how to develop and deploy an ERC721 (NFT) smart contract.

Tech Stack

• Alchemy
• OpenZeppelin
• Remix
• Ethereum Goerli

1. Open Zeppelin

First of all, let’s go to OpenZeppelin, then on the Wizard tab, click on ERC721. All NFTs are ERC721 tokens.

We want our tokens to be Mintable, have Enumerable and URI Storage as the features. Name your token as per your wish and also enter the desired symbol.

2. Remix IDE

Now, click on the Open in Remix button in the upper right-hand side corner. It will open the Remix IDE and will have the code written. You can change the compiler version to 0.8.4.

Here, is the updated code, be sure to change it in Remix. We need to remove the Ownable part of the solidity program as we want that people other than ownable can also mint our NFTs. Also, we have added MAX_SUPPLY as 5000 so that the NFTs are limited.

3. Alchemy- Create app

Next, we need to sign up for Alchemy. Click on Create App and name your app whatever you wish. I named it Namzz nft. Then click on View Key and copy the HTTPS address.

4. Metamask- Add Network

Further, we need to create our network on Metamask by clicking on Add Network. Name the testnet as per your wish. I named it Namzz Goerli testnet. Paste the https URL that we copied from alchemy. Add ID as 5 for Goerli and enter ETH as the symbol. Click on Add Network button.

You can get some test Ethereum from the Goerli faucet. Here is the link. Login with your Alchemy account and enter your address. You will get the test ether within seconds!

5. Compile and Deploy the NFT Smart Contract on the Goerli Testnet

On Remix, go to Deploy and select Injected Web3 as the Environment. It will get connected to you network automatically. You will get the address of your metamask along with the balance ether. Select your smart contract in the Contract Tab and click on Deploy.

It will show a metamask popup asking you to pay the fees. Click on Confirm and wait for 10 seconds. It will add the deployed contract in the Remix!

6. Formatting the NFT Metadata

For OpenSea to pull in off-chain metadata for ERC721 tokens, the contract will need to return a URI pointing to the hosted metadata. To find this URI, OpenSea, Rarible and other popular marketplaces will use the tokenURI method contained in the ERC721Uristorage standard.

Head over to OpenSea developer doc and copy the metadata structure. Paste it in json online editor and edit the description according to your wish. According to the OpenSea documentation, the NFT Metadata should be stored in a .json file and structured as follows:

7. Creating and Uploading the Metadata on IPFS

Now that we have a brief understanding of what will be contained in your NFT metadata, let’s learn how to create it and store it on IPFS- Inter Planetary File System.

Go to Filebase, and make an account there. Verify your email and login. Now, we need to click on create bucket, the bucket name being in small case letters, I named it namzz-nft. Select IPFS and create. Now upload the image you want to make as the nft. Click on the image and copy the IPFS Gateway URL. Enter this link in the image section of JSON file.

Save the JSON file in the name of metadata.json and upload it on Bucket in FileBase.

Click on your file and copy the IPFS CID.

8. Mint Your Goerli Testnet NFT

Then head on over to Remix. Orange methods are methods that actually write on the blockchain whereas Blue methods are methods learning from the blockchain.

Click on the safeMint method dropdown icon and paste your address and the following string into the uri field:

ipfs://<your_metadata_cid>

Clicking on transact will create a Metamask popup prompting you to pay the gas fees. Click on “sign” and go on minting your first NFT!

Enter your address in the balanceOf button and enter your address. Run it — it should show you have 1 NFT.

Do the same with the tokenUri method, inserting “0” as the id argument — it should display your tokenURI.

Great! You just minted your first NFT! 🎉 Now it’s time to move to OpenSea to check if the metadata is read.

9. View the NFT on OpenSea

Navigate to testnets.opensea.io and log in with your Metamask wallet. Then click on your profile picture, you should see your newly minted NFT there. If the image is not yet visible, click on it, and click on the “refresh metadata” button.

Sometimes OpenSea has hard times recognizing testnet metadata — and could take up to 6 hours to see it.

Congratulations, you have successfully created, modified, and deployed your first smart contract. Minted your first NFT, and published your image on IPFS!

Thanks for reading.

Create NFTs Using Solidity With Open Zeppelin and Publish It on IPFS was originally published in Better Programming on Medium, where people are continuing the conversation by highlighting and responding to this story.

Let’s go 👇

1.ADDRESS — a string of characters that represents a wallet that can send and receive cryptocurrency It is like a real-life address or email. Each address is unique and marks the location of a wallet on the blockchain. Example: 0x71C7656EC7ab88b098defB751B7401B5f6d8976F

2. ALPHA — early investment advice or intel ‘Where It Happens podcast by @sahilbloom and @gregisenberg is the best source of alpha for success in 2022.’ @NDerewlany ‘The real alpha in crypto is buy and hold.’ @RyanSAdams

3. WEN — when

4. WAGMI — we’re all gonna make it

5. NGMI — not going to make it

6. IRL — in real life

7. SER — sir, a polite way of disagreeing with someone

8. APEING IN — buying into an NFT project, perhaps without due diligence ‘In just 1 day @KarafuruNFT already had 1 sale over 100 ETH, floor rise to 4.2 ETH after reveal … getting DMs from big names that are apeing in.’ @eddyiskongz

9. BLOCKCHAIN — a decentralized database that maintains a data list that cannot be manipulated Examples: Bitcoin, Ethereum, Solana, Polkadot, Terra and others.

10. BTFD — buy the f’ing dip When people are running around and selling because a crypto price is dropping hard, this is the time to buy. “Solana is under $100. BTFD.”

11. BUIDL — build useful stuff When crypto is dipping, building useful stuff proves that its valuable. “Eth is crashing, we need more developers to BUIDL.”

12. CRYPTO — cryptocurrency A digital currency that is secured by cryptography, which makes it nearly impossible to counterfeit or double-spend. Cryptocurrency runs on a blockchain. Ether is the token that runs on the Ethereum blockchain, Solana runs on Solana etc.

13. DAPP — decentralized app Any practical application of blockchain and/or cryptocurrency. dApps can take the form of mobile games, communications platforms, and social media sites. @CryptoKitties is a dApp game that allows users to buy and sell virtual cats.

14. DAO — decentralized autonomous organization The structure for a company or organization in web3. In a DAO everyone can have ownership and be involved in decision-making.

15. DEGEN — degenerate, a gambler, someone really shooting at all chances, taking extremely high risks Refers to people who invest in digital assets like NFTs or cryptos without doing due diligence. Can also mean any person getting in on web3 investments.

16. DEFI — decentralized finance DeFi is the movement within crypto that removes the control banks and institutions have on money, financial products, and financial services. It eliminates the fees that banks and other financial companies charge for using their services.

17. DYOR — do your own research Don’t depend on others. Do your own research before you invest in cryptos, buy NFTs, or get involved with a community. ‘DYOR. That’s it, that’s the tweet.’ @TrustWallet

18. FIAT — a term for government issued currencies, like the U.S. dollar or Japanese yen Fiat is used to describe any currency controlled by a central authority. Cryptocurrencies with their decentralized design, serve as a counterpoint to traditional fiat currencies.

19. FLOOR — the lowest price available for an NFT in a collection The minimum amount you need to buy one. This is an important metric to look at when evaluating an NFT project. “The floor price on Bored Apes went from 0.5 ETH to 2 ETH in 3 days!”

20. FOMO — fear of missing out An emotion of that drives you to make wild decisions. When an NFT project rises sharply in value and investors buy in because they are afraid to miss the boat. ‘Getting off Twitter y’all are giving me FOMO already.’ @YaOnlyLivvOnce

21. GAS — the fee required to conduct a transaction or execute a contract on the Ethereum blockchain Gas prices are denoted in small fractions of ether called gwei. ‘Gas limit’ refers to the maximum amount of gas that you’re willing to spend on a particular transaction.

22. GENESIS BLOCK — the first block in the blockchain, used to initialize the crypto-currency

23. GENESIS DROP — an artist’s or project’s first ever NFT drop ‘Wormhole, my genesis drop on @SuperRare is now live.’ @KalebsCode

24. HODL — holding on for dear life or holding on despite loss When a crypto tumbles, loyal investors urge each other to think long term and HODL — not sell their tokens. ‘Crypto will make a new generation called billionaires.’ #SHIB #HODL @DuncanGray

25. MAXI — maximalist Someone who believes that one particular crypto is the only viable one by far and predominantly invests in it. ‘I have my fair share of bitcoin, but I’m more of an Ethereum maxi.’ @mcuban

26. MINT — the creation of a new NFT token Minting is the process of turning a digital file into a crypto collectible or digital asset on the Ethereum blockchain. ‘981 NFTs already minted 419 left Mint your BabyShiba NFT now’ @babyshibacoinio

27. NFT — non fungible token An NFT is a digital asset that confers ownership of a virtual good, such as a piece of digital artwork or online collectible. Most cryptocurrencies are “fungible”, meaning there is no meaningful distinction between one coin and another.

28. P2E — play to earn, as in video games like @AxieInfinity

29. PFP — picture for proof or profile pic like @crypto_coven

30. POAP — proof of appearance protocol from @poapxyz

31. GMI — gonna make it

32. PROBABLY NOTHING — often used sarcastically to mean the opposite. It probably is something. Can also refers to the Probably Nothing podcast by @alexisohanian and @TZhongg

33. SATOSHI — 1. a reference to Satoshi Nakomoto, the anonymous founder of Bitcoin who vanished shortly after his creation, leaving the project to be decentralized and collectively managed 2. a unit of exchange, the smallest part of a bitcoin. 1 Satoshi = 0.00000001 ฿

34. SHILL — the act of unsolicited endorsing in public You’re shilling if you’re promoting an NFT you’re dropping or a coin you’ve invested in. “Shill me your best NFT under 1 ETH.”

35. WALLET — a place to store crypto tokens HOT WALLETS are online and easily accessible but also more susceptible to hackers. COLD WALLETS store digital assets off-line, making them secure from bad actors but difficult to trade. Examples: @MetaMask, @phantom, @CoinbaseWallet.

References: https://coinmarketcap.com/alexandria/glossary

Please share this article to help more people find their way into web3. I write regularly about the blockchain, web3 and smart contracts.

See ya! Till then, BUIDL!

Join Coinmonks Telegram Channel and Youtube Channel learn about crypto trading and investing

Also, Read

• 5 Best Crypto Trading Terminals | Best DeFi Apps
• Best Online Casinos | Binance Review | BitMEX Review
• Sparrow Exchange Review | Nash Exchange Review
• Best Crypto Trading Bots in the US | Changelly Review
• Earn Passive Income Using Crypto Arbitrage In India
• Godex.io Review | Invity Review | BitForex Review
• 5 Best Free Cryptocurrency Charting Tools

40 Terms to make you SMARTER in the Web3 Space was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.

It often leaves one confused that in which project should one invest in. The purpose of this post is to educate the potential investors in evaluating the projects in the Web3 ecosystem.The goal over here is to help provide a framework to make informed decisions for the investors.

One needs to view it from a practical perspective as there are also some projects that may be of low quality and not have any backend value to them. Also, we need to see whether the project is adding value to the customers’ daily life or not. It should be smooth and with a good unique selling proposition.

We need to have a Framework so that we can evaluate the Web3 projects and be saved from the faulty projects too. The basic framework that I would suggest includes the features listed below so that we can analyse the Web3 project and invest in it.

• USP
• Community Engagement
• Developer Side
• COCOMO strategy
• Tokenomics
• Risks

Rapid innovation makes us feel that the time and investment to be done in the right place is really essential. It is important to understand various mechanisms and terminologies so that we can move ahead in the Web3 space with ease. This article discusses the framework and gives suggestions on how to evaluate each feature related to the project.

Let’s dive right in-

USP

The first and the foremost thing that we can evaluate on the basis of is the project’s unique selling proposition (USP). It should make value to the customers. The project should mean something and have something unique about it. The case should not be like just for the sake of being in Web3, they have associated a project to the blockchain. For example, the USP of stable coin TerraUSD is that it works on the basis of LUNA. It has a unique selling proposition. So, one needs to be fascinated about the product or service that is being offered by the project in the Web3 space. Also, the project may be similar to the one in the Web2 space, so why would we invest in something that has already be done before? It should provide a better flow and have an advantage over the Web2 version of the product. Thus, the unique selling proposition should be strong enough in order to attract us to invest in the project.

Community

Most of the times, we see that a project is successful when it already has a strong community before even launching. This is a key factor to check the community strength over on the Web3 project’s social media profiles, the most popular one being Twitter. All in all, I would look at the project’s social media that includes Discord, Twitter and Reddit. Community plays a very important role as they are the ones that make a project know about their faults, validations and feedbacks honestly. Also, one more important law that comes into being with Community is the Metcalfe’s law.

It gives the importance of Network effect and thus, increases the probability of a project being more successful and thus, more profitable. Social media helps to track the response towards negative feedback of the community and their problem solving in real time. We can look at the number of followers and engagers and thus, get to know about the engagement of the community for a particular project. Also, we can look at the tags, all the good projects are either referred in the popular pages of the accelerators and thus, we can get to know about it through social media.

Developer Side

The best thing about most of the Web3 projects is its open source nature. The projects being open source help the investors to get access to the code and the development strategies. It helps to know the working of the teams and see the foundation stage of the project. It also helps to know about the project’s ideas and innovation game. We can get an idea on how the team would be and also the team’s working in the development stage in the project planning. We are also able to see the changes done every now and then and get to know about the updated versions from time to time. For example, Solana Labs have their open source project on Github, here is the link. The Buidlers are really hardworking or not can be known by viewing their open source project!

COCOMO Strategy

When we want to know about a project in terms of mathematics, we can evaluate it by a model called the Cost Constructive Model (COCOMO). It uses regression modelling and helps to get to know about a project’s productivity. The basic COCOMO model applies to three types of projects:

• Organic — Small teams with good experience working with flexible requirements
• Semi-detached projects — Medium teams with mixed experience working with more rigid requirements
• Embedded projects — Projects under tight constraints

1. Product attributes

• Required software reliability
• Size of application database
• Complexity of the product

2. Hardware attributes

• Run-time performance constraints
• Memory constraints
• Volatility of the virtual machine environment
• Required turnabout time

3. Personnel attributes

• Analyst capability
• Software engineering capability
• Applications experience
• Virtual machine experience
• Programming language experience

4. Project attributes

• Use of software tools
• Application of software engineering methods
• Required development schedule

Thus, it is a good model that tells us the features for the project to assess on. We can use COCOMO whenever the project seems to be correct with the first three factors of USP, community and developer tech side.

Risks

While evaluating a project, just take a look at various factors that may pose a risk to the project itself. The three basic risks involved are:

1. Price Fluctuations in Tokens of the Project

The tokens may directly go up once they are launched, but the other way round is also possible. So, we need to take care of the community and development factors to get a confidence into the price fluctuation and stability scenario. The token should be able to gain momentum and also withhold it, only then can it be known as a good project.

2. Smart Contract Risks

The smart contracts are known to be secure but at the end, they are made by the human developers. Thus, they may have bugs that may issue vulnerabilities when the project is launched. So, a good practice would be to check and verify that the smart contract for a particular project is audited and verified to be free of bugs such as Reentrancy. Also, if somewhere there is an integration of two or more smart contracts, it may pose a risk of attack because it would lead to vector collisions of array. So, the web3 project should have a verified smart contract whose testing has been done.

3. Lending and borrowing risks

When the Web3 project is related to a lending platform, an important risk to take into account i the lending strategy. What if all of the community members start to withdraw the tokens once unlocked, then the price would just crash. So, the Web3 project should have some safety mechanism in order to stay protected from such kinds of events.

Conclusion

Web3 is a huge place for the projects to come. The most important thing is not to have fear of missing out (although, it may be difficult sometimes!). Each project is unique in its own way but the quality lies on the factors of the framework discussed above. I believe that the above lens can help investors make a bet on the long-term future of Web3 projects while avoiding the pitfalls of a speculative ones.

Disclosure: I am not a financial advisor in any way. Please do your own research while investing in the projects. The above opinions are my own. It is to share my idea of investing for the potential investors.

Contents distributed by Learn.Block6.tech

👉 Telegram — Fresh ideas

👉 Twitter — Latest articles

👉 LinkTr.ee

Deep Dive: Evaluating Web3 Projects was originally published in BLOCK6 on Medium, where people are continuing the conversation by highlighting and responding to this story.

Vending machine is an interesting thing to get started with the basics of Solidity. So, we are going to write a code in Solidity that will help us build a smart contract for a Donut Vending Machine!

Let us start off by going to Remix which is an IDE for the language of Solidity. It is used to write Smart Contracts for the Blockchain. In terms of implementation coding for ERC-20 tokens, the six basic coding functions are:

1. getBalance
2. purchase
3. restock

Remix

Remix IDE allows developing, deploying and administering smart contracts for Ethereum-like blockchains. It can also be used as a learning platform.

It is very user friendly and helps beginners to know about all the basic things required to deploy the contract.

First, we start by creating a file in the name of say, vending_machine.sol in the File Explorers tab.

Code

Solidity 0.6.8 introduces SPDX license identifiers so developers can specify the license the contract uses. SPDX license identifiers should be added to the top of contract files. The following identifier should be added to the top of your contract:

We now need to write the version of solidity that we are going to be using:

The next step is to create a contract. In Solidity, we write a smart contract by typing the keyword contract followed by the name of the contract. Let us proceed by naming our contract VendingMachine. So, write:

We will write our further code in the curly brackets as all the functions and variables would be a part of it. Next, we define the variables that will be a part of the contract that is:

1. address public owner;
2. mapping(address => uint) public donutBalances;

Here, address refers to the wallet address of the owner and mapping refers to a keyword where we can map one type of variable to be associated with another.

So, donutBalances is a mapping where number of donuts will be associated with the address. Next, we go on to create a constructor that will be automatically executed when deployment will occur on the Ethereum Blockchain.

The creation of first function is getBalance(). It is used to get the balance of donuts that are left in the Vending Machine.

donutBalances[address(this)] represents the balance of the donuts that are associated with the current Ethereum account.

Function that is next is called restock(). It is restricted only to the owner and thus, the keyword comes as require.

The donutBalances gets updated when the amount is entered by the owner to be incremented.

The last function is the purchase function that actually helps the person or the customer to purchase a donut.

The price for one donut that we entered is 0.5 ETH. Thus, the msg.value should be equal to the amount multiplied by 0.5 ETH.

Also, the Vending machine should have the number of donuts entered by the customer. When both of these requirements are fulfilled, then the donutBalances of this address gets deducted by the amount.

Also, the msg.sender’s donutBalances will get updated by the number of donuts purchased.

Finally, the code is done. Now we, proceed to compile it. The Remix automatically compiles the code and shows if there is some error or not. Then, we go ahead to Deploy the code.

Make sure that we are in Deploy and Run Transactions tab. Then the environment to be selected is Javascript VM. The contract should be our Vending Machine contract.

Deploy it now, and you will get the deployed contract as follows. Also, now whenever the customer has purchased the donuts, the donutBalances gets updated when the function is called.

Now, when we call the restock function from the owner’s account, the getBalance gets updated to 100 again!

The Vending Machine is ready and so is our Smart Contract.

If you have any issues, you can refer to the GitHub link:

GitHub - Namyagupta/vending_machine: A vending machine contract

Building Smart Contracts for a Donut Vending Machine was originally published in Better Programming on Medium, where people are continuing the conversation by highlighting and responding to this story.