Naqqash Abbassi is the CTO of mydost.ai, where he focuses on developing AI-powered solutions to streamline and automate repetitive tasks for SMBs, specifically in Accounts Payable (AP) Automation. With a background at Tech Data and Arrow Electronics, he supported partners and customers through digital transformation, working with major vendors like Intel, Microsoft, AWS, and IBM. Naqqash also brings five years of research experience in renewable energy, focusing on power electronics and energy storage. Passionate about Generative AI, he frequently shares his expertise in Document AI, Small Language Models (SLMs), Multimodal Large Language Models (MLLMs), Agents, Quantization, and AI Product Development.
Naqqash, can you share the story behind the founding of mydost.ai and what inspired you to focus specifically on Accounts Payable Automation?
A colleague, now my co-founder and CEO, once approached me with the idea of creating digital co-workers. After brainstorming, we concluded that there was a strong business case for Small and Medium Businesses. As we interviewed potential customers, we discovered that the finance sector, particularly Accounts Payable (AP), had the greatest need for optimization through AI and automation. Recognizing the significant time consumption and potential for improvement in AP, we decided to focus our efforts on automating Accounts Payable processes.
How do you see AI-powered solutions, like those developed by mydost.ai, transforming the landscape of small and medium-sized businesses in the next five years?
With the continuous progress in AI and automation, conventional businesses are already feeling the impact. In the next five years, we will see most medium-sized businesses adopting cloud-based SaaS platforms, which will significantly reduce operational complexity and burden.
Given your extensive experience with companies like Intel, Microsoft, AWS, and IBM, how have these collaborations influenced your approach to AI product development at mydost.ai?
In my previous roles as a Solution Architect working with technologies from major industry players, I gained extensive experience and understanding of how to execute such projects effectively. I learned the importance of focusing on the use case and solving problems for the end user, rather than getting caught up in the hype of the technology.
Can you elaborate on some of the most significant challenges you faced while transitioning from roles in large corporations to co-founding and leading a startup?
Being part of a large corporation comes with its challenges; you need to create impact, be a subject matter expert, and work as a team player. On the other hand, as a founder and CTO, you need a diverse set of skills that extend beyond technical expertise to include strategy and management. Many times, I found myself in situations where I had to learn something new and tackle it for the first time. Additionally, the level of responsibility and ownership in a startup is pretty different from that in a corporation.
In your research on renewable energy, particularly in power electronics and energy storage, what key lessons did you learn that are applicable to your work in AI and automation today?
As a research engineer, I focused on designing and implementing stochastic estimation techniques-based algorithms for estimating the State of Charge (SOC) and State of Health (SOH) in energy storage systems. Additionally, I managed substantial amounts of data related to cell aging research. This experience laid the foundation for my transition into the world of the Internet of Things (IoT) and AI.
With the rapid advancements in Generative AI, how do you envision the role of AI in shaping the future of work and the necessary skills for upcoming professionals?
We are already noticing how generative AI, with tools like ChatGPT, is impacting our productivity. Companies are integrating these technologies into their existing products to provide a seamless experience. It won’t be long before we have our own copilots working seamlessly on our devices, catering to our needs based on our usage and data.
Document AI and Small Language Models are areas of particular interest to you. Could you explain how these technologies are integrated into mydost.ai’s solutions and their impact on business efficiency?
Data extraction from financial documents is at the core of our work. We continuously improve our technology and stay updated with the latest research in this area. With the rise of Small Language Models, we anticipate that smaller, locally implementable models will effectively solve specific tasks. This approach is particularly relevant to our use case, where different agents collaborate, each solving a subset of a larger problem.
What strategies do you employ to stay at the forefront of AI innovations, especially in the rapidly evolving domains of Multimodel Large Language Models and AI quantization?
As mentioned before, we ensure that we stay up to date-with the latest research. Where applicable, we also strive to incorporate these advancements into our own products to take full advantage of emerging technologies.
Reflecting on your journey, what personal and professional milestones are you most proud of, and how have they shaped your leadership style at mydost.ai?
When we started DOST, there were several unknowns, and it was uncharted territory. I learned to be more pragmatic and not to approach challenges solely from a technical perspective. By hiring, managing, and mentoring people, I developed leadership skills that enable me to motivate and help my team members grow. At the end of the day, perseverance and consistency are crucial; without them, it is almost impossible to survive as a startup founder.
Looking ahead, what are your long-term visions for mydost.ai, and what key milestones or projects do you hope to achieve in the coming years?
We aim to become the leading player in our domain. Our goal is to create a platform for CFOs that will serve as a comprehensive financial stack and a copilot, assisting them in making better financial and strategic decisions
