Building vs. Buying AI Solutions: Lessons from Optimal Blue's Journey

In the ever-evolving landscape of technology, companies are increasingly faced with the crucial decision of whether to build or buy AI solutions. At Optimal Blue, Seever Sulaiman, CTO, shared valuable insights from their journey, highlighting the considerations and strategies that shaped their approach to AI implementation.

The Build vs. Buy Dilemma

Optimal Blue's journey in AI began long before generative AI became the buzzword. With a history of using machine learning models and predictive analytics, the company has been ahead of the curve in leveraging AI technologies. However, the advent of generative AI about a year ago presented new opportunities and challenges.

One of the first steps in their AI journey was to evaluate whether to build their own solutions or partner with established providers. This decision hinged on several factors:

1. Core Business Relevance: Is the AI product or feature core to your business? For Optimal Blue, certain AI functionalities, like sentiment analysis within a CRM tool, were deemed essential and therefore required a deeper, in-house understanding.

2. Support and Maintenance: Building in-house means long-term maintenance and updates. Optimal Blue, with its large engineering team, opted to build organically, ensuring they could support and evolve their AI solutions.

3. Control vs. Partnership: Maintaining control over AI systems was crucial for Optimal Blue. While they partnered with Microsoft for tools like Copilot and OpenAI's GPT, they kept their data private, avoiding public LLMs to protect sensitive information.

4. Training and Education: A significant part of their journey involved educating their team. Partnering with experts, such as Microsoft architects, provided hands-on training and helped them build robust AI frameworks.

Implementing AI: A Structured Approach

Optimal Blue's implementation of AI followed a structured approach. They developed a framework focusing on three key pillars: productivity, product development, and security.

1. Productivity: The integration of tools like Copilot and natural language processing (NLP) techniques enabled their engineering team to work more efficiently, automating routine tasks and accelerating innovation.

2. Product Development: They deployed AI assistants and incorporated generative AI features into their products, enhancing customer experiences and providing valuable insights.

3. Security: Protecting data was paramount. Optimal Blue implemented a framework to ensure that none of their data or code was shared with public models, maintaining control and confidentiality.

For Startups and Smaller Companies

For smaller companies or startups considering AI, the key takeaway is to focus on education and training. While partnerships with major players can provide resources and expertise, it's essential to develop an internal understanding of AI technologies. Experimentation, guided by experts, can also lead to innovative solutions that align with your business needs.

Looking Ahead

The future of AI in applications is bright. Generative AI is becoming an integral part of software solutions, and companies like Optimal Blue are leading the way in integrating these technologies. As AI continues to evolve, the strategic alignment of data, products, and AI will be crucial for businesses to stay competitive and innovate effectively.