Designing the UX of tomorrow with LangChain

Riding the wave of the next big paradigm shift

Fire, farming, steel… the iPod. Humans have invented some cool stuff over the years.

And every now and then, a new technology pops up and changes everything—a civilization-wide paradigm shift that alters the course of human history, if you want to be dramatic about it. 

Ships crossed the oceans, connecting peoples from worlds away. But these were soon replaced by airplanes, deepening our global interconnection.

Or consider how In the span of roughly 100 years, we went from music being a strictly live event, to the phonograph, cassette tapes, CDs, MP3s, and now streaming, giving us instant access to nearly all the music that’s ever been recorded.

And of course, the revolutions of cloud storage (buh-bye save icon) or the smartphone, which connected us with anyone, anywhere in the world at any time, and put the sum total of human knowledge in the palm of our hands.

These are not mere technological changes—they fundamentally altered how we interact with the world and with each other.

As UX professionals, we have a duty to keep up with these changes to help us deliver the best possible experiences to our users. 

So as we stand at what appears to be the precipice of another such upheaval—the AI revolution—what are some of the technologies that we can incorporate into our UX practice to help us ride the wave and not just get swept away by it?

I propose that simply learning to use generative AI technologies like ChatGPT isn’t enough. 

UX professionals should take it one step further and learn what's possible when we incorporate more advanced AI technologies. 

For example, LangChain.

What the heck is a LangChain?

LangChain is an open-source framework created by Harrison Chase in 2022—and I realize that may mean basically nothing to some of you. Fair enough. 

So how about this: LangChain is software that makes it easier to build apps that use large language models (LLMs) like GPT-4. Let’s say you want to build an app that leverages ChatGPT to help users blocking the internet for your kids id their laundry basket is filled. LangChain makes it easier to do that.

Now, for us UXers, this doesn’t mean we need to become developers. But understanding a tool like LangChain will help us understand what kind of digital experiences it’s now possible to give our users. And that is a very important thing, especially if you want to stay ahead of the competition.

If you’re interested, hear what LangChain’s creator has to say about its capabilities.

LangChain vs ChatGPT

We’ve already established that LangChain is a tool that helps you leverage technologies like ChatGPT to create apps and experiences. But let’s do a quick comparison just to clarify.

ChatGPT:

Ask it a question, get an answer. It might be brilliant, it might be bullshit, but you’ll get an answer.

LangChain:

Ask it a question, get a whole lot of nothing. That’s because LangChain is a tool to help build things—it’s not a chat app.

… … 

ChatGPT:

Gives you general-purpose conversational AI that can be used out-of-the-box for nearly anything you can think of.

LangChain:

Gives you the building blocks to create custom AI applications for nearly anything you can think of. 

… … 

ChatGPT:

You’re the end-user.

LangChain:

You’re making things for the end-user.

… …

ChatGPT:

Super easy. If you can type, you can use it.

LangChain:

Super not-so-easy. If you can type, you’re still gonna need to learn how to build apps to use it. But that’s OK, because…

Remember: the goal here isn’t necessarily to learn to make apps with LangChain. The goal is to understand what’s possible with it so you can design the user experiences of tomorrow.

Let the devs handle the code while we designers explore what’s possible with this new technology.

You can see more about what exactly LangChain is here

The UX possibilities of LangChain

Chatbots that don’t suck

A tool like LangChain can take our chatbot game from zero to chatbot hero. A LangChain app can have a much more sophisticated understanding of context compared to plain ol’ gen-AI like ChatGPT. This means more sophisticated chatbots with a deeper understanding of the nuances of a conversation. So users will have a more natural and less frustrating experience. 

Tech like LangChain will help push chatbots over that hump so they become truly useful by seemingly understanding the user at the same level as a human customer support rep. 

Lightspeed summarization and analysis

A LangChain app can summarize a massive amount of information at incredible speeds. This could revolutionize how we present information to users, allowing for dynamic content adaptation based on user preferences or time constraints.

If you’re not familiar with Slack’s AI recap feature, it creates customized recaps of use-selected channels and groups. With tech like LangChain and the right LLM, you could easily add a feature like this to give users summaries of long conversation threads, lengthy documents, or even entire chat histories.

Check out this case study on building a text summarization app with LangChain.

Hyper-personalized recommendations

Products like Spotify, Netflix, and Amazon already offer sophisticated personalization, but they use similarities between users to make recommendations. Basically they just categorize users. 

But tech like LangChain can bring it to a higher level with advanced natural language processing and data integration capabilities for hyper-personalized experiences. It means the AI will learn about you as an individual, and not just put you into a bucket with similar users, for eerily accurate suggestions.

Semantic search

LangChain can power advanced search functionalities that understand the intent behind a user's query, rather than just analyzing the keywords. 

This could dramatically improve how users find information within applications. Maybe products like Slack, Jira, Confluence, and others now offer enhanced search functionality that grasps context and intent, making it easier to locate specific messages, files, etc. within the product.

So basically, the UX search is changing before our eyes

Multi-modal Interactions

Finally, leveraging LLMs can give users a seamless experience when using both traditional inputs (clicking, typing, etc.) and more natural language-based ones. 

For example, Adobe Firefly lets users generate images by simply describing what they want in plain language. But then, users can use traditional design interfaces tools to refine the results. This powerful multi-modal approach can make users feel like Superman for an incredible UX experience.

Embrace the wave…

… Or be left behind. That’s how I see it. Staying on top of the latest technological developments help us build better digital products and experiences. 

And to be sure, LangChain is just one of many frameworks being built to leverage the power of generative-AI. There are many other new and exciting advances coming down the pipe.

So, what’s your take on leveraging AI for better UX? Any ideas for how a technology like LangChain might be used to “wow” our users with a mind blowing experience?

Let me know!