Personal and professional history might not repeat, but it certainly rhymes. I’m thrilled to join the team at HumanFirst, and reconnect with a team of founders I not only trust, but deeply admire.
The skill of Prompt Engineering has been touted as the ultimate skill of the future. But, will prompt engineering be around in the near future? In this article I attempt to decompose how the future LLM interface might look like…considering it will be conversational.
Personal and professional history might not repeat, but it certainly rhymes. I’m thrilled to join the team at HumanFirst, and reconnect with a team of founders I not only trust, but deeply admire.
Google launched Vertex AI on 18 May 2021 at Google I/O & it seems like the product has faired well considering all the recent advances in LLMs and Generative AI
Generative AI & Predictive AI
LLMs have two key applications: Generative and Predictive. Generative AI, also known as Gen AI, has been widely adopted and developed for its use-cases, and I think this is in part because of its straightforward implementation. Prompt Engineering is the go-to method for Generative AI, particularly in the realm of language.
Google Cloud started at the opposite spectrum of LLMs, by focusing first on predictive prior to generative.
However, there is a challenge with the predictive side of things, as one needs to train a model based on specific training data, and must ensure a high level of accuracy, especially if the model is being implemented for a specialised domain.
I find it fascinating that Google Cloud has only just recently released semi-publicly available generative systems, such as Bard and the Google Generative AI App Builder, which can be used to create LLM Apps and Generative Apps (Gen Apps).
Additionally, Google Research and the University of Washington have collaborated and developed the LLM-based Prompt Chainer, as seen below.
Read more about the PromptChainer application here.
Predictive Is Important
Intents are commonly understood by those in the Conversational AI and Chatbot circles; however, they can be simply thought of as classifications. This concept of classification is an essential part of any predictive AI system.
Intents are just another term for classes and the process of classification.
Classification involves assigning pre-defined labels to classes and providing a set of training data to teach a model. This model can then be used to predict the classification of any user entry, be it a single phrase or a longer piece of text.
More On Vertex AI
Vertex AI offers four classification tasks when selecting the TEXT tab:
Single label &
Multi-label text classification,
Entity extraction, &
Sentiment analysis/classification.
These tasks are part of the various classifications available in Vertex AI, which also include images, tabular data, and video.
Generative AI
Below is an extract from Google, and how they see the future of Generative AI:
Google Cloud will launch a range of products that infuse generative AI into our offerings, empowering developers to responsibly build with enterprise-level safety, security, and privacy. This journey starts today with the introduction of two new technologies:
Generative AI support in Vertex AI gives data science teams access to foundation models from Google and others, letting them build and customise atop these models on the same platform they use for homegrown ML models and MLOps.
Generative AI App Builder allows developers to quickly ship new experiences including bots, chat interfaces, custom search engines, digital assistants, and more. Developers have API access to Google’s foundation models and can use out-of-the-box templates to jumpstart the creation of gen apps in minutes or hours.
I’m currently the Chief Evangelist @ HumanFirst. I explore and write about all things at the intersection of AI and language. Including NLU design, evaluation & optimisation. Data-centric prompt tuning and LLM observability, evaluation and fine-tuning.
Subscribe to HumanFirst Blog
Get the latest posts delivered right to your inbox
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
.webp)
.svg)
.svg)
%20(42%20x%2024%20in)%20(1).png)
.png)
.png)
%20(280%20x%20208%20px)%20(4032%20x%202304%20px)%20(1).png)
