Get the "AI for Small to Mid-Size Businesses" white paper for strategies, tools, and more.
Thanks for subscribing!
Oops! Something went wrong while submitting the form.
Icon Rounded Closed - BRIX Templates

AI Experiments: Progress and Learning from Our Work in AI

October 8th, 2024

Weekly Experiments in Building with AI

The Helm team gets together each week with a shared excitement to delve deeper into the world of artificial intelligence. These sessions are pivotal steps toward embracing the future of technology within our work and how we shape the world moving forward.

Charged with excitement, we're sharing our learning along the way.

Web shift > Mobile shift > AI Shift

Our session underscored that AI represents a significant platform shift, much like the rise of the internet or mobile applications in previous decades.

Businesses are increasingly recognizing the necessity of AI integration to stay competitive, much like businesses needing a website in the late 90's.

Clients and friends are regularly inquiring about AI solutions, with various organizations expressing interest to move fast. There's also notable attention from investors, indicating strong market confidence in the value AI brings.

Learning 1: Build On a Strong Foundation

One developer on our team updated us on progress with a new mobile app we're developing. The primary focus has been on establishing a solid foundation by enabling chat messaging functionality and deploying the app for initial testing. [In one day!]

Users can now send and receive messages, and real-time push notifications alert them to new interactions. We’re strategically designing the app's architecture to seamlessly incorporate AI features next, but a solid ground to work from is paramount.

By selecting APIs compatible with AI capabilities, we're ensuring that the transition to intelligent features like conversation summarization and categorization will be smooth.

Our next steps involve optimizing message retrieval to reduce API consumption and beginning to integrate AI elements that will enhance user interactions within the app.

Learning 2: Harness Company Data with AI

Another team member worked on creating an interface that references our internal project data from the tools we use daily, like Asana and Harvest. The goal is to enable queries using real-time project information, which aims to transform how we access and utilize our data.

A script was developed to import project management data into our custom knowledge base. This “AI assistant” or “Digital twin” has the potential to answer questions about team workloads, project statuses, and more, providing immediate and actionable insights.

However, challenges are inevitable! Were encountered issues with data limitations, causing the AI model to exceed context size limits and crash. We're currently troubleshooting these issues, aiming to optimize data handling to stay within limits. Resolving this will pave the way for more complex queries and a more robust assistant.

What's Next?

It’s clear that careful planning and strategic actions position companies at the forefront. At Helm, our next focuses include:

• Continuing to build robust foundations in our applications to readily support these AI integrations.
• Resolving technical challenges to optimize AI functionalities within our working tools.
• Working with clients and friends to help them understand the technology and the use cases that apply to their businesses.

Our continued journey into AI is filled with exciting possibilities and challenges. By experimenting and laying the groundwork now, we're creating innovative, intelligent solutions that transform our own workflows and those of our clients.

More to come - stay tuned.

Take the next step towards your AI future.