Generative Op by ComexiasLabs - Your Generative AI Workspace

Project Name: Generative Op

Project Track: Web3

Team Name: Comexias Labs

Team Members: @jnlewis

Devpost Project Link: https://devpost.com/software/genop-8cw13g

Demo Video Link: https://youtu.be/5ulk2EkuLfo

Project Website: https://www.generativeop.com

Project Test Instructions

  1. Visit https://www.generativeop.com
  2. Click on Sign In on the top right
  3. Once signed in, click on Open Your Workspace on the top right

Open Source on GitHub: https://github.com/ComexiasLabs/generative-op-extensions

Smart Contract Links: N/A (This platform interacts with the Blockchain but does not possess its own smart contract)

Generative Op: Your Generative AI Workspace

Project Goal: Our goal is to unlock the full potential generative AI technology to businesses by providing a seamless experience to explore, create, and manage AI-driven workflows tailored to specific business needs - without the need for any setup or hosting AI models.

The inspiration for Generative Op (short for Generative Operation) came from the growing need for accessible AI tools that can adapt to various industries and use cases. As AI technology rapidly evolves, we saw an opportunity to bridge the gap between complex AI setups and everyday business use. We wanted to create a platform that offers powerful generative AI capabilities and also provide the flexibility to customize and integrate external tools into their AI workflows.

About the Project

Generative Op (short for Generative Operation) is a cutting-edge SaaS platform designed to empower businesses with the full potential of generative AI. Built with flexibility and scalability in mind, Generative Op allows users to explore over 100 expertly crafted AI prompts across various industries or create their own custom prompts tailored to their specific needs. The platform supports seamless integration with popular apps, tools, blockchains, and APIs, providing a truly customizable experience.

With Generative Op, users can manage their prompt libraries, develop extensions, and take advantage of developer-ready APIs for smooth integration into their existing workflows. We also prioritize AI safety, offering customizable safeguards to ensure privacy, truthfulness, and security in all AI-driven tasks. Whether you’re looking to supercharge your productivity, streamline your creative processes, or explore new AI possibilities, Generative Op is your go-to platform for AI-powered innovation.

*Supported AI models: Anthropic Claude, Meta Llama, Amazon Titan, Mistral AI Mixtral, Cohere Command, AI21 Labs Jurassic

Integration with TRON Blockchain

Generative AI models today are pre-trained and can only answer according to the knowledge that it has learned. What if you want to ask something that requires real time integration with TRON? For example, asking for the current balance of a wallet, or analysis on the most recent transactions and blocks?

Here is where integration becomes important. At Generative Op, we have built an open source extension for TRON that users can integrate with their prompts.

How it works

  1. When you create a prompt, you can choose to activate the TRON extension, which gives you additional options, such as Mainnet/Testnet, wallet and contract addresses.
  2. Then you can choose which APIs you want to access, including wallet and account details, contract details, and block information.
  3. Now you can write your AI prompt to ask questions on these data.
  4. When the AI is generating a response to your prompt, Generative Op will integrate with TRON and fetch the necessary data in real time for the AI model so that it has the most up to date information.

Features

Access industry leading AI models

  • With support for over ten large language models from various AI companies

Presets for any use case

  • Instantly access 100+ expertly crafted prompts tailored for diverse use cases, ready to deploy and use.

Your own prompt library

  • Create, manage, and customize your own dynamic prompt library to suit any project with flexibility and control at your fingertips.

Extensions and customizations

  • Integrate generative AI with your favorite app and blockchains, or build your own extensions for a tailored experience.

AI safety and privacy

  • Implement customizable safeguards for privacy, truthfulness, and security, ensuring your AI operates safely and ethically.

Developer integrations (Available soon)

  • Connect with your applications with built-in API and webhooks, designed for smooth and scalable integration.

Business Model

Summary: Free and publicly available

Generative Op will run on a freemium model where majority of features will be made free with set quotas and with the availability for increased quotas and team management for business and enterprise users.

Technical Details

Architecture Overview

Web Application

  • The frontend Web Application is developed in React using the NextJS framework. The authentication, analytics and core APIs are developed in NodeJS.
  • The web applications is hosted on Vercel.

APIs

  • There are two custom built APIs powering Generative Op:
    1. The core API which manages all platform operations
    2. The generative AI API which streamlines model inferences and enable real time streaming
  • Both of these APIs are hosted in AWS Lambda and developed in NodeJS express.

Database

  • The database in use is DynamoDB on AWS and used to manage users’ prompt library and configurations.

User Identity Management

  • We leverage on Clerk for user identity management, authentication and authorization.

Blockchain

Contract Interactions/Deployments

  • TronWeb and Web3Js libraries are used for all contract interactions, including deployments. This will be available via an extension and is currently in development.

Block Explorers

  • We rely on data from block explorers such as TronScan API for our TRON integration extension. This is already available under the TRON Scan extension.

Screenshots

17 Likes

Welcome to session 7 of hackathon
I’m particularly intrigued by the potential applications of GenOp in different industries.
Could you please elaborate on some specific use cases or examples of how businesses might benefit from this platform.

2 Likes

welcome team, trying to understand but it is not going. Please kindly summarize everything for me. What are you building and jhow can this product be of help to me?

3 Likes

Welcome buddy. What happened to this your project.
Yinbox

2 Likes

Welcome to Hackathon Season 7, after a thorough read I see this supports multiple generative AI models and provides a user-friendly workspace where businesses can explore pre-built AI prompts or create custom ones and that’s interesting cause of the general Generative AI adoption.
How would you handle situations where the pre-trained generative AI models fall short, especially in providing context-specific or real-time information?

2 Likes

Hi Okorie, thanks for bringing this up. One of the main barrier for businesses to adopt AI in their products is the complexity and cost to build such capabilities, so this the gap that GenOp wants to bridge. One example that people might see more frequently is in the travel agencies and sites like TripAdvisor, where they use generative AI to produce custom tailored itineraries. However, TripAdvisor is a well establish organisation and have the resources to do this, but there are also smaller businesses across various industries that may not have engineering or AI capabilities to build advanced features to remain competitive. On GenOp, once a user logs in there are over 100 use cases they can play with to give them ideas on what they can build for their business.

Hi Prince, good to see you around. I understand this project is not easy to relate to because of the target audience. Since it is a B2B saas platform, so the target are businesses instead of individuals.

There is a fierce race for businesses to leverage on generative AI today, you can see this across many industries and use cases, from education, travel to recruitment and marketing. Eg: when you apply for a job online, the company may use gen AI to summarize your resume. The problem is, it is difficult and expensive to build these AI features in their systems, especially when they have to host and run their own AI models. GenOp give them a no-code platform where they can just configure the use case they want, and then its made available to them to use whether directly on GenOp or through an API. In short, GenOp makes it easy for any business to use AI directly in their products without hiring an entire team of specialist engineers.

Hi Gordian, thanks for bringing this up, I started Yinbox as an individual but have moved on to a more structured working arrangement now as ComexiasLabs with additional help and advisors. Internally we have discussed about Yinbox and one thought that we agree with is that there are now telegram apps that can do similar things and is a better user experience. We’re still considering our next steps, but for our current focus (at least in this quarter) is to continue growing BlockFabric and to launch the MVP of GenOp to test market reception.

Alright, so what is the latest growth level for BlockFabric S6

2 Likes

oh OK I get it, thank you

2 Likes

Welcome to Grand hackathon S7
I think it’s still mixing up on my end trying to understand mostly your base or target area to impact, I grabbed a little about business but got confused where I saw Tron related stuffs, I would like to understand it more clearly especially your targeted audience

2 Likes

Welcome to hackathon season 7. Can you elaborate more on the AI feature, what exactly is it built up on i.e what model? Or is it an AI being made by this project?

Hey manfred, thanks for the question. This is one of the challenges with generative AI currently, giving it access to information it doesn’t have. For example in TRON, you can’t ask ChatGPT to find real time information on the blockchain. How we solve this here is to build an extension that links GenOp to TRON Scan’s API, where whenever we ask an AI a question, GenOp will fetch all the relevant data from the API and then pass it to the AI so that it has the latest information.

1 Like

Hello, thanks for highlighting this confusion. It seems we have some work to do on our end to make the project more understandable, and I think a lot of this will clear up once we have the demo video out. Our target audience are people who want to use generative AI capabilities in their own apps, or use it directly in GenOp. While we currently made only text based AI models available, our next plans is to introduce image generation to enable more possibilities. So for example if someone wanted to build an NFT with generated images and deploy it directly to TRON, it’s just a manner of configuration with no code. But this is just a small example. For now, our TRON integration include giving AI access to TRON scan API so people can ask questions like how they do with ChatGPT, but it will be able to get real time data from TRON. Example like: “tell me what this wallet balance is”. We’re still in early stages and this is intended to be a B2B model,. so our target users would be new or existing start ups wanting to integrate AI into their products.

2 Likes

Awsome idea for market nowadays

1 Like

@jnlewis
Wow :flushed: this is amazing but given the extensive use case library on GenOp,
how can businesses effectively leverage this resource to identify and prioritize the most impactful AI applications for their specific needs and goals.

1 Like

Welcome to the Hackathon of season 7, I am still behaving fascinated when it comes to AI things, please tell me are there data backup tools that you are going to be using, thank you

1 Like

Thanks for taking out the time to clarify.

1 Like

Thanks for the clarification, would be waiting for demo video, all the best as you continue to build

2 Likes

Welcome to hackathon S7 Sir, Since your target audience are businesses using GenOp is it going to be for free? and again did you have any tutorial that this business people can learn to understand it better?
And I saw library please explain to me how this got connected with library.
Thank you.

1 Like