Learning Library

← Back to Library

Watsonx.ai Prompt Lab Overview

Key Points

  • Watsonx.ai is an enterprise studio that unifies generative AI and traditional machine‑learning tools, letting users build, train, tune, and deploy models tailored to specific business problems.
  • In the Prompt Lab, users can craft prompts from scratch or use sample prompts for tasks like summarization, sentiment analysis, or question‑answering, choosing from a curated catalog of foundation models—including IBM’s Granite series and third‑party models such as Llama 2—and adjusting parameters and guardrails to control output quality and safety.
  • Completed prompts can be instantly shared via a one‑click API command, saved to a Jupyter Notebook for further testing, or refined iteratively with different models, decoding techniques, and parameter tweaks.
  • The platform also includes a Tuning Studio for fine‑tuning foundation models on proprietary data and a no‑code AutoAI environment for building traditional predictive models, enabling seamless collaboration between application developers and data scientists within a single workspace.

Full Transcript

# Watsonx.ai Prompt Lab Overview **Source:** [https://www.youtube.com/watch?v=swPBNKKPK0E](https://www.youtube.com/watch?v=swPBNKKPK0E) **Duration:** 00:05:28 ## Summary - Watsonx.ai is an enterprise studio that unifies generative AI and traditional machine‑learning tools, letting users build, train, tune, and deploy models tailored to specific business problems. - In the Prompt Lab, users can craft prompts from scratch or use sample prompts for tasks like summarization, sentiment analysis, or question‑answering, choosing from a curated catalog of foundation models—including IBM’s Granite series and third‑party models such as Llama 2—and adjusting parameters and guardrails to control output quality and safety. - Completed prompts can be instantly shared via a one‑click API command, saved to a Jupyter Notebook for further testing, or refined iteratively with different models, decoding techniques, and parameter tweaks. - The platform also includes a Tuning Studio for fine‑tuning foundation models on proprietary data and a no‑code AutoAI environment for building traditional predictive models, enabling seamless collaboration between application developers and data scientists within a single workspace. ## Sections - [00:00:00](https://www.youtube.com/watch?v=swPBNKKPK0E&t=0s) **Untitled Section** - - [00:03:05](https://www.youtube.com/watch?v=swPBNKKPK0E&t=185s) **No‑Code Model Building & Synthetic Data** - The passage explains how AutoAI enables rapid, no‑code creation, training, and deployment of predictive models, and how Watsonx.ai’s synthetic data generator can fill data gaps by automatically producing large, realistic datasets for model training. ## Full Transcript
0:00Watsonx.ai is an enterprise studio that puts generative 0:04AI and traditional machine learning at your fingertips, empowering you 0:09to build, train, tune and deploy 0:12AI models targeted for your specific business needs. 0:16Let's begin our tour of watsonx.ai in the Prompt Lab. 0:20Here you can build a prompt from scratch or select one of the sample prompts 0:25corresponding to common generative AI tasks, including applying summarization 0:30to condense long form content, like an earnings call, into short descriptions. 0:35Or deploying insight extraction and classification 0:39to determine critical insights, like the sentiment of a customer review. 0:44Each sample prompt defaults to the pre-trained foundation 0:47model most appropriate for the task at hand. 0:50Our curated list features foundation models of different sizes and architectures, 0:55including third-party open models like Llama 2 0:58and IBM-developed models like the Granite series. 1:01The models in the Granite series help businesses 1:04build and scale generative AI and we offer certain contractual intellectual property 1:09indemnity protections for these IBM-developed AI models. 1:14You can experiment with different models from the list 1:16and adjust their parameters. 1:19Let's say you want to create a question-answer application 1:23for your company's financial statements. 1:25By clicking the appropriate sample prompt, you can quickly start 1:28prompt engineering in either a freeform or structured 1:33format. 1:34In either format, you can provide natural language instructions 1:37and examples, which helps your model uncover patterns. 1:41You can also add AI guardrails, 1:43which can help prevent your model from receiving or generating 1:47harmful, abusive and profane language. 1:50Okay, let's test your prompt. 1:55Prompt engineering is an iterative process. 1:58You can adjust your prompts, test different models 2:04and tinker with decoding techniques and parameters. 2:07Once you've built the most effective prompt for your use case, 2:11it's easy to share across your organization. 2:13With one click, you can retrieve the API command and send it to a developer. 2:19You can also save the prompt to a Jupyter Notebook. 2:22From there, a data scientist can do additional testing. 2:26Watsonx.ai also includes a Tuning Studio, 2:30enabling your enterprise to harness its proprietary data. 2:34Simply select a foundation model, tuning method and task. 2:38Then upload a dataset. 2:41Once your model is tuned, you can start using it in the Prompt Lab. 2:46Watsonx.ai also supports building traditional machine learning workflows 2:50and models, facilitating collaboration between application developers 2:54and data scientists in a single workspace. 2:58Let's imagine you're looking to create a machine learning model 3:01designed to process and forecast upcoming customer insurance claims. 3:05A data scientist can then quickly create a traditional predictive model that forecasts 3:11the claims amount by connecting to a repository 3:14of historical claims data in a project. 3:17With AutoAI, you can create a machine learning model 3:20from scratch in a no-code environment. 3:23Now you can pull in the training data file, 3:26set the model type, adjust 3:29its parameters and specify the range of options. 3:33You can run experiments in AutoAI. 3:36The most promising pipelines and models are identified on a leaderboard. 3:41You can save the winner as Python code 3:43or immediately deploy it for inferencing. 3:47A developer can use the model endpoint 3:49to integrate the new model with their code and UI. 3:53And just like that, you've built an app. 3:56Now, let's say you didn't have enough data to effectively train your ML model. 4:01In this case, you can leverage watsonx.ai’s synthetic data generator 4:06tool to address your data gaps, whether stored in a database or as a data file, 4:12you can upload your existing data, anonymize columns, 4:16and select how many rows of synthetic data 4:19you want to generate - up to 2.1 billion. 4:23You can also select the type of data file you want output. 4:26Watsonx.ai takes it from there, automatically learning your data’s 4:31distributions and relationships. 4:33In short order, it creates a robust synthetic dataset 4:38that reflects your original training data. 4:40Watsonx.ai is part of the watsonx AI and data platform, 4:45comprised of three core components and complemented by several AI assistants designed 4:51to help you scale and accelerate the impact of 4:54AI with your trusted data across your business. 4:58The core components include: a studio for new foundation models, generative AI 5:03and machine learning; a fit-for-purpose data store 5:06built on an open data lakehouse architecture; and an AI governance toolkit 5:11that accelerates responsible, transparent and explainable AI. 5:16To learn more, start a free trial or reach out 5:19to an IBM representative.