Today we invite Olga Khryapchenkova, Phd, AI Product Manager at NIO on ‘Behind the Mic’ to share her valuable insights

Dr. Olga Khryapchenkova serves as the Lead Experience Manager for NIO’s in-vehicle AI system, NOMI, overseeing the development of a user-friendly voice interface. Her role involves integrating the latest technologies in conversational AI and generative AI, including automatic speech recognition, natural language understanding, skill development, text-to-speech technologies, and large language models.

Driven by user feedback and insights from UX research, Olga collaborates with NOMI’s international R&D team to translate these findings into new skills and features. These innovations aim to enhance driver safety and improve the overall user experience with NIO vehicles, fostering a sense of joy and convenience in car usage for NIO users.

We had the privilege of speaking with Olga, a seasoned expert in AI product management with extensive experience in natural language processing and generative AI. Here’s what she shared with us.

Q1. What inspired you to pursue your career in AI product management, and what were the key steps in your journey?

After years of experience in product localization and research, followed by a couple of years in dialog development, transitioning to product management in natural language processing and later generative AI felt like a natural progression. Being involved from the earliest stages of product planning—identifying user needs, shaping solutions, and collaborating with diverse stakeholders—is what makes this role so fulfilling.

Q2. How have you utilized emerging technologies like AI and machine learning to build user-focused and innovative products?

While AI can optimize processes, my primary focus has been on integrating it directly into products, in order to deliver user-centric solutions. For speech assistants, this includes wake-word detection, accurately recognizing user requests through automatic speech recognition and natural language understanding, as well as generating appropriate, engaging content with the help of LLMs. Conversational AI has tremendous potential to make interactions between users and products—and also with brands—both efficient and emotionally engaging. At the heart of this work lies the ability to bridge cutting-edge technology with real user needs. A cornerstone of successful product development is continuously collecting user feedback and effectively managing expectations.

Q3. What are some of the most significant challenges you've faced in managing AI-driven product development, and how did you address them?

Addressing emotional aspects, such as designing human-centric AI, determining appropriate levels of AI anthropomorphizing, and managing the amount of system feedback. Delivering lasting value requires not only meeting user needs but also ensuring cultural appropriateness and relevance. These challenges have been both complex and rewarding, requiring a thoughtful, user-focused approach at every stage of development.

Q4. What role does user feedback play in shaping your AI products, and how do you ensure it's effectively incorporated into your development process?

At NIO, we integrate user feedback throughout the product development process using a closed-loop system. We gather insights at key stages, whether to validate product requirements with real user input, draw inspiration from user ideas, or measure feature satisfaction. Feedback is collected through various channels, including our app, NOMI —our AI voice assistant—and direct user meetings, ensuring a comprehensive approach to shaping and refining our products. This helps us improve our product in the right direction with every release.

Q5. What strategies do you use to deliver seamless and meaningful user experiences in AI-driven products?

Our AI voice assistant, comprehensive user research, the above mentioned feedback loop system, the design-driven approach, the personalization, and the commitment to innovation collectively guide the product development process. Operational excellence, necessary for it to happen, is achieved through close communication and collaboration with both technical and non-technical stakeholders. Additionally, we prioritize the emotional aspect, considering how users should feel while interacting with the product.

Q6. Can you share an example where data analytics or user behavior insights helped identify opportunities for improving your AI product?

During the early stages of our product launch in a new region, we discovered that the amount of proactive feedback from our AI voice assistant was too high for users, reflecting different UX expectations across cultures. User feedback and research helped us identify this preference, allowing us to make the right adjustments in a subsequent release.

Q7. Could you highlight a success story where your work in AI product management led to impactful outcomes for users or businesses?

One of last year’s highlights was the successful launch of Generative AI capabilities. For use cases where scalability and unpredictability are key, the traditional intent-response systems can fall short. Collaborating with exceptional technical teams (Product, Development, Deployment, QA, etc.) and non-technical teams (Product Marketing, Corporate Communications, Legal, etc.), we set clear targets and delivered impactful features, driving innovation and value. We faced lots of challenges, and it made us as the organization and our product stronger. Now we are building upon this experience, continuously improving it.

Q8. What frameworks or methodologies do you rely on to prioritize features while ensuring alignment with user needs and organizational goals?

The Kano model helps us categorize features into must-haves, performance features, and delight features, while also recognizing indifferent and reverse needs, enabling us to prioritize based on their impact on user satisfaction. We prioritize features by combining qualitative user feedback for actionable insights, quantitative data to identify usage patterns and gaps, aligning with our design and product principles for strategic direction, and conducting a thorough evaluation of resource requirements for feature development.

Q9. How do you foster collaboration with cross-functional teams—like engineering, data science, and design—to create cohesive AI products?

Early alignment on feature goals, development process, testing process and feature communication; maintaining efficient communication through streamlined channels; ensuring clear task delegation with timelines; and promoting ongoing collaboration across teams to drive progress and accountability

Q10. How do you balance pushing the boundaries of AI innovation with ensuring your products remain accessible and user-friendly?

It’s about trusting UX research and user feedback to guide priorities, while remaining agile and ready to pivot when needed. It’s about being brave and curious, yet attentive to detail and open to change when necessary.

Q11. What measures do you take to maintain ethical AI practices and ensure transparency in your products’ interactions with users?

On one hand, we collaborate closely with our legal team to stay aligned with the latest regulatory frameworks and recommendations. On the other hand, we prioritize clear, transparent communication with users and host frequent events where they can ask questions and gain insights into our product development processes.

Q12. How do you stay ahead of trends and continuously learn about advancements in AI and product management to deliver user-centric solutions?

I regularly follow AI news portals like “The Decoder” and similar platforms to stay updated on key areas: research breakthroughs, foundation model advancements, partnerships, AI regulations, cloud technologies, MLOps, hardware developments, applications, and – last but not least – AI ethics and philosophy.

Conclusion

In her insightful interview, Dr. Olga delves into the complex world of AI product management, where cutting-edge technology converges with user-centered design and ethical practices to deliver impactful solutions. She highlights the challenges of balancing innovation with usability and emphasizes the critical role of user feedback in shaping value-driven and culturally relevant AI experiences. By adopting advanced methodologies, fostering cross-functional collaboration, and staying ahead of industry trends, Dr. Olga exemplifies the continuous learning and adaptability required to build successful AI products. Her approach not only addresses present user needs but also anticipates the evolving demands of a rapidly advancing technological landscape, ensuring that AI-driven solutions remain relevant, effective, and inclusive.

“Thank you for tuning in on Behind the Mic, where experts like Dr. Olga share their experiences and ideas to inspire innovation and create user-focused solutions in today’s digital world.”