Insight & Analysis | Automation

AI-Powered Productivity Gains in Firmware, Insurtech and Beyond: Q1’25 Roundtable

May 14, 2025

AI-Powered Productivity Gains in Firmware, Insurtech and Beyond

Insights from THL’s Quarterly AI Roundtable

Key Takeaways:

To help our partner companies evaluate the business use cases for, and accelerate real-world adoption of AI, THL hosts quarterly roundtables for the CTOs, CIOs, and tech leaders across its portfolio to share wins, challenges, lessons learned, and advice.
During the Q1 2025 session, two THL portfolio companies presented case studies: AMI showcased how its internal AI assistant, AMILiA, dramatically improved engineering productivity, while Hexure shared its journey implementing GitHub Copilot to modernize development workflows and achieve efficiency gains of over 40%.

Accelerating product roadmaps. Automating tedious coding tasks. Streamlining repetitive workflows. Retrieving information in an instant. These are just some of the ways generative AI is helping companies boost productivity. The technology shows immense promise across industries and disciplines, and we’ve only just scratched the surface on its potential.

Recognizing both the promise and complexity of AI adoption, THL Partners (THL) established a quarterly roundtable series in early 2024 to bring together CTOs, CIOs, and other tech leaders across its portfolio. These virtual roundtables are designed as a forum for open collaboration on all things generative AI.

THL Vice President Alex Sabel, who leads the firm’s research efforts, moderates the discussions. “Our portfolio roundtables bring together people from completely different sectors and experiences, creating a trusted environment to share wins, challenges, and ideas,” he said. “It’s been powerful to see companies that might never cross paths exchanging insights on common tech hurdles and lessons learned.”

The Q1 2025 session focused on “AI for Productivity,” featuring engaging presentations from portfolio companies AMI and Hexure. Both companies offered real-world examples of how they are embedding AI into their workflows to drive measurable impact. “What’s truly exciting about these pilots is that we’re seeing technologists harness AI to turbocharge their effectiveness,” said Jagjit Singh, Director, THL. “It’s about helping developers focus on strategy and impact.”

Here are some key highlights.

Scaling Internal Knowledge with AMI’s “AMILiA”

American Megatrends, Inc. (AMI) is a pioneer in BIOS, server management, and open-source firmware, supporting computing platforms, laptops, servers, and embedded devices worldwide. The company has decades of proprietary knowledge—roughly 80,000 BIOS-related documents accumulated over 20-30 years, to be more exact—that they wanted to harness to make their engineers’ jobs easier. Generative AI was a no-brainer.

Led by Sudan Ayanam, AMI’s Head of Technology and Architecture, the team developed “AMILiA,” an AI-powered internal assistant designed to drive major productivity gains. Taking its name from “AMI’s Linguistic Assistant,” AMILiA serves as an advanced internal knowledge management tool that makes decades of technical insights easily searchable and accessible. They leveraged proprietary and open-source large language models and a secure cloud-hosted vector database to create an accessible, searchable database that customer support engineers could use to resolve inquiries in minutes rather than hours. When a query comes in, engineers search AMILiA for similar past issues, enabling them to respond and resolve problems significantly faster.

In a pilot involving 600 support queries, AMI reduced total resolution time from 300 hours to just under 10 hours. Queries that once took between 30 minutes and an hour – even more – can now be answered in just one or two minutes.

“We were able to save about 280 hours,” Ayanam said during his presentation. “And this is on a smaller set [of support queries]. We believe that rolling out this to a wider set of audience, we’ll be able to even save more.”

The project did come with challenges. AMI discovered that approximately 60% of its legacy data was unusable due to incomplete records, prompting a rigorous data-cleaning effort. Protecting proprietary information and maintaining data separation between product lines also required sophisticated namespace segmentation and strict role-based access controls. With AMILiA’s strong foundation and early success, AMI plans to expand the platform across its entire engineering organization and eventually offer it to external customers, potentially transforming how firmware support is delivered industry-wide.

AMI has also been developing a source code generation assistant to further accelerate engineering workflows. By using AI to analyze hardware design documents and datasheets, the tool can automatically generate key platform configuration files—early testing shows up to 90% accuracy. AMI envisions eventually integrating this capability directly into developers’ IDEs, creating a Copilot-like experience specifically tailored for firmware engineering.

Modernizing Development Workflows with Hexure’s Copilot Strategy

After AMI’s presentation, Warren Perlman, CTO of Hexure—a sales automation platform for insurance and financial services industries—shared the company’s experience implementing GitHub Copilot to improve developer productivity. Facing an aging codebase and developers using outdated IDEs, Hexure needed to modernize its technical infrastructure before successfully integrating AI tools.

Following THL’s 2024 Coding Assistants Pilot initiative, Hexure made the strategic decision to migrate its engineering teams to Visual Studio 2022, enabling seamless Copilot integration. The transition paid off: within a year, 79 of Hexure’s 87 developers were actively using Copilot.

The Copilot adoption led to notable gains. Developers experienced a 40%+ boost in efficiency, particularly in areas like code commenting, unit test creation, and code refactoring. Copilot’s success was further amplified by a focus on cleaner, more declarative code across the team.

Migrating legacy environments was no small task. Hexure invested heavily in updating its infrastructure and built internal dashboards to monitor Copilot usage, fostering accountability and encouraging broader adoption. As Perlman shared, “Now that everybody’s on VS22 and we’re starting to see some of the benefit, we are seeing a tremendous improvement in delivery. I’m seeing greater than 40% in efficiency improvement.”

As an added benefit, Hexure expanded its use of AI tools beyond engineering, leveraging Copilot across operations and administrative functions to enhance productivity company-wide.

Building the Future: What’s Next for THL’s Quarterly Roundtables

The success of THL’s AI roundtables underscores the immense value of shared knowledge. Tech leads from several portfolio companies—Bynder, BazaarVoice, and Fortna, to name a few—asked questions throughout the roundtable and planned follow-up conversations to delve deeper into specific topics and questions. By creating a forum for open dialogue and live peer feedback, these gatherings help portfolio companies surface new ideas, avoid common mistakes, and build momentum faster.

Looking ahead:

Next Roundtable: The next session will dive deep into the evolving role of “AI Agents,” or multi-step, task-completing agents capable of significantly expanding automation possibilities.
In-Person AI Summit: In June, THL will bring its entire portfolio of tech, healthcare, and financial services companies together in-person for a major AI summit co-hosted with AWS. This event will include CEOs, CTOs, CPOs, and product leaders, expanding the conversation from operational improvements to full product innovation strategies.

On THL’s AI roundtables, Mark Benaquista, Managing Director, THL noted, “The breadth of our portfolio allows for unexpected cross-pollination of ideas. When a firmware leader like AMI and a digital asset management platform like Bynder can sit at the same table debating best practices, the collective knowledge of the portfolio leaps forward.”

THL remains committed to fostering these forums, strengthening its community of innovators, and helping portfolio companies stay at the forefront of the AI revolution.

Visit www.THL.com to learn more, and stay tuned for more insights after our upcoming sessions—the AI journey is just getting started.more insights after our upcoming sessions—the AI journey is just getting started.

—

Mark Benaquista, Managing Director, SRG
mbenaquista@thl.com

Jagjit Singh, Director, SRG
jsinghthl.com

Alex Sabel, Vice President, TABS
asabel@thl.com

^{This piece includes statements provided by certain executives of THL’s portfolio company, some of whom may also be limited partners in the THL funds. None of the executives were compensated for providing these statements. As a result of the ownership structure of these portfolio companies and the executives’ investments in the THL Funds, a conflict of interest exists because the executives have an incentive to make positive statements about THL and their experiences with THL to maintain the goodwill of THL. Any statements reflect the subjective views of the speaker and THL has not independently verified such information and makes no representation or warranty, express or implied, as to the accuracy or completeness of the information contained herein.}