This summer, I had the opportunity to join Fiducia AI as a Software Engineering Intern through the CITRIS Workforce Innovation Program (WIP). CITRIS WIP is an eight-week University of California internship program that helps students build in-demand skills via real-world projects. Some background info on Fiducia, they're an early-stage startup revolutionizing sports fan engagement with AI-driven experiences. I tackled projects at the intersection of AI and security. From building a malware scanner to prototyping a grapevine disease detection tool, the experience was packed with learning. In this post, I'll reflect on what I built and what I learned.
What I Built
AI-Powered Malware Scanning Module
One of my primary projects was developing a malware scanning module for user image uploads. This feature is crucial in any platform that accepts files to ensure bad actors can't inject viruses or malicious content. I integrated ClamAV (an open-source antivirus engine) with a Python FastAPI backend to scan images on the fly. Technically, this meant containerizing ClamAV's scanning daemon and updating its virus signature database (via FreshClam) so our service could detect known malware by matching file contents against thousands of virus signatures. I built an API endpoint that intercepted image uploads, passed them to ClamAV for scanning, ultimately returning secure uploads to pass to the LLM. We also wrote unit tests to simulate file uploads and ensure infected files are correctly caught. This project taught me how signature-based malware detection works and reinforced the importance of secure coding practices. On the business side, the scanner helps protect our platform and users, preventing potentially dangerous uploads from ever reaching our systems. This is a big win for fan safety and trust.
Grapevine Fungal Disease Detector
Another exciting project was the Fungal Detection Activation, a prototype feature using AI to identify grapevine fungal diseases from images. This was quite different from the sports domain, but it showcased Fiducia's versatility and an agricultural use-case for our tech. I crafted the prompts for the LLM to interpret the vision model's output, for example, describing the spotted patterns and asking the LLM to conclude if it's powdery mildew or black rot, and to do so in plain language. I also contributed to the UI/UX design of the mobile interface, ensuring the results were displayed clearly with an option for multilingual support. We had the system output the diagnosis and advice in English and Spanish to make it accessible to more users. It was deeply rewarding to see an AI feature go from idea to a visual demo. It struck me that AI productization isn't just about model accuracy, it's about the user experience around the model output and making it actionable.
Additional Contributions
Beyond these major projects, I wore many hats as a startup intern. I wrote unit tests to keep our code reliable, assisted in developing backend APIs in both Python and Java (exposure to Spring Boot taught me how robust enterprise APIs are built), and helped deploy new features to our cloud environment. Working across different tech stacks (from FastAPI endpoints to Spring Boot microservices) in a short time sharpened my adaptability. I also practiced prompt engineering in various contexts, learning how subtle wording changes can significantly alter an AI's response. These technical contributions made me a much stronger developer by the end of the summer.
Lessons Learned
My internship wasn't only about coding, actually it was rarely coding. It was an immersive experience in how a startup operates. Some key takeaways:
Aligning Tech with Business Value
I learned that every engineering task should trace back to a business goal. In our team meetings, we constantly discussed why a feature mattered to users or clients. For example, the malware scanner wasn't just a neat security add-on, it was protecting brand integrity and user trust, which has real business value. Seeing how my mentor (also CEO) evaluated feature priorities taught me to always ask, "How does this solve the user's problem or improve the product's value proposition?"
Collaboration, Communication & Documentation
I participated in daily Scrum calls and even sat in on a client-facing meeting. Initially, I was nervous to speak up, but I gradually gained confidence updating the team on my progress. I also discovered the importance of clear documentation and reproducibility in a fast-moving project. By writing setup guides and API docs, I made sure others could run my code and that my work lived beyond my internship. Good documentation isn't just a "nice-to-have", it enables effective collaboration and faster onboarding for new team members. I now appreciate how sharing knowledge (through comments or demos) is as vital as writing the code itself.
Cross-Functional Teamwork
Working with designers, engineers, and even non-technical stakeholders was eye-opening. I saw how feedback loops from different perspectives improve the final product. For instance, our designer's input on the fungal detection UI helped us simplify the user flow, and an engineer's code review pushed me to handle edge cases I hadn't considered. I learned to welcome feedback and ask clarifying questions, which is a habit that improved my work quality and will serve me well in any team setting.
Personal Growth
This was a remote internship, which required self-direction and discipline. I set daily goals and learned to unblock myself by researching or asking the right questions, rather than waiting passively. Over eight weeks I became much more independent as a developer. I also learned to balance taking initiative with seeking mentorship. My mentor (Ganesh) encouraged me to propose solutions first, then he'd guide me with adjustments. This approach boosted my problem-solving confidence while still providing a safety net for learning. By summer's end I felt more confident not just in writing code, but in designing solutions and explaining my thinking.
Real-World Impact of AI
Perhaps the most inspiring lesson was seeing the real-world impact of the technologies I worked on. I helped build a malware scanner that could safeguard users in a digital community, and an AI-driven tool that might help farmers detect crop diseases early. And of course, Fiducia's core mission which is using AI to make sports fan experiences more interactive and engaging has the potential to change how we all enjoy games. This breadth, from sports entertainment to agriculture and cybersecurity, showed me how versatile AI technology can be. It reinforced my belief that by pursuing AI and software engineering, I can contribute to solving diverse, meaningful problems.
Gratitude & Next Steps
I'm extremely grateful to the Fiducia AI team for welcoming me and making this internship so enriching. A huge thank-you to Ganesh, my mentor and manager, for his patient guidance, technical insights, and for challenging me to grow. His feedback and high expectations pushed me to write cleaner code and think like an engineer, not just a student. I also thank the other engineers and designers who took time to answer my questions and include me in important discussions.
Additionally, I want to thank the CITRIS Workforce Innovation Program organizers for facilitating this internship. Programs like CITRIS WIP are invaluable in bridging classroom learning with industry experience, and I appreciate the workshops and support they provided throughout the summer. Although this will be the last cohort of WIP, I'm incredibly grateful to be apart of this fantastic community.
As I head back to campus this fall, I carry with me new skills in Python, Java Spring Boot, FastAPI, and prompt engineering, as well as a deeper understanding of how AI products are built and delivered. This internship has affirmed my passion for software engineering and AI, especially building systems that are secure and have positive real-world impact. In the future, I aim to dive further into AI in production (maybe even exploring the cybersecurity side of AI) and perhaps work at the core of AI and social good.
As I wrap up this summer, it's important for me to think about my future. In the short term, I plan to apply these lessons in my coursework and personal projects such as: writing more tests, documenting thoroughly, and considering the user impact of everything I build. Long term, whether I end up at a startup or a larger company, I feel better equipped to contribute from day one thanks to this experience.
Concluding, my summer at Fiducia AI was a period of tremendous growth. I shipped features that stretched me technically, learned from inspiring mentors, and saw firsthand how innovative tech is aligned with business strategy. Most importantly, I left with even more excitement for the field of AI and a clearer vision of the kind of engineer I want to become. I'm proud of what I accomplished in those eight weeks and excited to keep building on that foundation in the years ahead. Thank you, Fiducia AI and CITRIS, for an unforgettable summer!