From Behavioral Sciences to a Career in Sensor Perception and AI

Every couple of years, I write a post about my career in which I review the past and consider pivots for the future. It is time to post an update for 2025. Because of AI, this year will be the most transformative.

I want to share some honest thoughts based on my journey and lessons learned.

Looking back over my 32-year career, I can see how high technology was always interlaced with my deep passion for cultural anthropology. Each past experience gently shaped my current path towards AI technologies.

Growing career skills through building on existing skills

My fascination with cultural anthropology and human behavior has been a lifelong passion, inspiring me to read countless books and contemplate human nature. I knew there was no career in anthropology, so I switched to computer science early on. However, the passion persisted to this day.

UI

My deep interest in behavioral sciences naturally guided me toward developing technologies centered around human experiences and emotional intelligence.

Developing the User Interface (UI) on the front page of Rolling Stone Magazine taught me the subtle importance of creating intuitive and appealing Human-Machine Interactions (HMI).

Web Crawlers

A couple of years later, building web crawlers to fetch massive amounts of Internet information, I gained experience in automation, parallelism, and efficiency. This experience opened my eyes to the power of systematically handling large amounts of data, a skill that became increasingly valuable as technology evolved.

Microservice Architecture

Later, I spent many years working on microservice architecture, another pivotal step highlighting the beauty of building scalable and adaptable systems of small, independent pieces. The principles of modularity and maintainability I learned here, such as enabling easier updates and better reliability, continue to inform my approach to technology today.

Android OS

Working on dozens of Android mobile applications and later leading the development of Android tablets for controlling precision farming tractors brought technology directly into practical, real-world applications, sensors, and implements. This role showed me firsthand how thoughtful tech can make life easier and more productive, even in complex environments like agriculture. My company, CNH, created a cab-less, self-driving tractor prototype, which inspired my next step. Watch the movie "Interstellar" from that period featuring CNH machines to see how we envisioned the future with tractors and drones.

Creating automotive infotainment systems at Harman was a critical experience. It combined consumer needs with stringent mass-scale automotive industry standards, reinforcing my belief that good technology works flawlessly and delights its users.

ADAS

Moving into Advanced Driver-Assistance Systems (ADAS), I encountered a deeper integration of machine learning and real-time processing. This experience was invaluable, teaching me about science, safety, and how technology and human behavior interact.

Sensor Perception of the 3D World

Currently, I work on advanced 3D LiDAR systems for perception technologies, a critical step in the evolution of robotics and automation perception. The world is multidimensional, not a flat screen; all robots and automated vehicles will have to build 3D Simultaneous Localization and Mapping (SLAM) and sensor fusion into their world. Working for my company gives me access to massive 3D data.

LLM and Agents

At the same time, the last few years have brought us LLMs and AI. I am passionate about local, secure, small language model clusters and multiple specialized agents. I am also obsessed with multiplying productivity, so I tirelessly build my “multitude” of AI agents.

Looking Forward

Looking ahead, I'm excited by robotics, specifically how clusters of small AI models and agents could provide intelligent, adaptive solutions in everyday robotics applications. These technologies will require minds with technical expertise, emotional intelligence, and creative adaptability.

Future job markets will undoubtedly change dramatically with increasing automation and AI. However, the core human skills of adaptability will remain essential. My friendly advice to anyone navigating their career today is simple: stay curious, embrace continuous learning, and use your diverse experiences to seek new insights. Creating that "multitude of personal AI models and agents" with many specialized skills will increase our productivity by 10x or even 100x. Maybe someday soon, I can share my experience with you, too.

Many people understandably worry about AI replacing human jobs, especially given near-term predictions of billions of AI agents becoming thousands of times smarter and faster than humans. While AI growth is unstoppable, our opportunities lie in ethically partnering with these intelligent systems and ensuring they enhance rather than replace human livelihoods. Reflecting historically, consider that 300 years ago, before 1760, almost everyone was a farmer, and nearly everyone walked or rode a horse. Horseless carriages would have been considered unnatural works of evil. Today, only 1.2% of people in the USA are farmers, and horses turn heads when spotted on a street.

Ultimately, a fulfilling life isn't about following a strict path but about enjoying the journey, learning from every experience, and gently allowing your past to inform your future. The future will be different, but being different is not necessarily bad. Perhaps in the future, a "Matrix" of cubicles we sit in for 10 hours a day will be considered abnormal and even cruel.