Uncovering GenAI - Chapter 2: Platforms, Tools and Community
Posted on 2025-05-21
GenAI
artificial-intelligence
The pace at which GenAI has exploded and gone from novelty to necessity is also reflected in the landscape of tools,
platforms, communities, breakthroughs, etc. The field is evolving so rapidly that staying up to date can easily feel
like a full time job. Hence, this blog focuses on the current state of the ecosystem to help readers get started.
GenAI Platforms
A GenAI platform refers to an underlying set of infrastructure that offers one or more capabilities - such as
using, developing, building, testing, and deploying GenAI applications. They can do that by providing pre-hosted
pre-trained model(s) or letting you bring or build model(s) of choice. You can then plug these models into your
GenAI application.
Let's classify these models under two types - Foundation Platforms and Deployment & Integration Platforms
Foundation Model Providers
These are companies or organizations that train and offer access to large language models (LLMs) like
GPT, Claude, Gemini, etc. They own and maintain the model(s). You can interact with these models via
chat UIs or via APIs. You typically have no control over the underlying model. Here is a simple example
where Gemini API is called, and gemini-2.0-flash model is selected to answer a simple prompt.
Deployment & Integration Platforms
Or, as I like to call them, the DIY-style platforms. These platforms help you build, host, and connect
GenAI models into your applications. These can offer open source models or allow you to deploy your own.
They additionally offer other tools and services to help in the process of building your application.
Platforms such as Hugging Face spaces, and AWS-Bedrock are some examples. Here is a screenshot from
Hugging Face spaces:
GenAI SDKs, Libraries and Tools
GenAI SDKs and Libraries help you go from idea to implementation - whether you are calling an API, chaining
model outputs, running models locally, or improving your development experience. There are lots and lots of
tools out there. You probably do not need to know all to get started, but some more popular ones to start with
are below:
| SDK, Libraries, and Tools |
Officially Supported Languages |
Description |
|---|---|---|
| LangChain | JavaScript/TypeScript, Python | This is a framework for developing applications powered by large language models (LLMs). |
| Semantic Kernel | Python, C#/.NET, Java | This is a model-agnostic SDK that empowers developers to build, orchestrate, and deploy AI agents and multi-agent systems. |
| OpenAI SDK | JavaScript/TypeScript, Python, C#/.NET, Java, Go | This SDK provides access to all OpenAI capabilities via OpenAI APIs |
| Anthropic SDK | JavaScript/TypeScript, Python, Java, Go, Ruby | This SDK gives access to Claude models via Anthropic APIs |
| Google GenAI SDK | JavaScript/TypeScript, Python, Java, Go | This SDK provides a unified interface to Gemini models through both the Gemini Developer API and the Gemini API on Vertex AI. |
| LLamaIndex | JavaScript/TypeScript, Python | This framework is used for building agentic generative AI applications that allow large language models to work with your data in any format. |
| LMStudio | - | LM Studio is a desktop app for developing and experimenting with LLMs locally on your computer. |
| Ollama | - | A command line utility to allow developing and experimenting with LLMs locally on your computer. |
GenAI Communities
It is always good to have a more localised community so do look out for communities of practice within your
organisation and in the region you reside. Local communities offer opportunities to network, ask questions
(suitable to your timezones), and spaces to show-and-tell. That said, mixing it with digital communities gives
you a well-rounded exposure to the lay of the land. Some popular communities are:
- Hugging Face Community - They publish articles, have social forums, and also have daily papers which can be insightful.
- OpenAI Community - A developer forum for those starting to integrate with OpenAI APIs. You can find some helpful resources if OpenAI is the platform of your choice.
- Reddit for LocalLLaMA - Subreddit to discuss Llama, the large language model created by Meta AI.
In Closing Thoughts
Awareness and curiosity are the best way to get started with the learning process of a new technology. GenAI is
no different, if it is an area of interest, then make yourself familiar with the basics and build a good
foundation (see the first blog written for the series). Try some passion projects out or solve at least one
of your day-to-day problems using GenAI and understand how integrations work. The above sections give you a
decent toolbox of your own to explore from. Stay tuned for some applications in the upcoming blogs for the
series.