Monesh Venkul Vommi
Senior Software Engineer at Virtue Group LLC (Senior Full Stack AI Architect)
About
I'm Monesh Venkul Vommi, a Senior Full Stack AI Architect currently at Virtue Group LLC. My career has been a journey from building apps and teaching 40,000+ students on Udemy to optimizing high-stakes banking systems at Capital One and ING. I specialize in moving AI from experimental notebooks into production-ready systems, with a heavy focus on RAG pipelines and graph-based backends. I’m deeply passionate about the 'build-first' mentality—I'd rather work with messy, real-world data than a clean tutorial set any day. Whether I'm reducing infrastructure costs or mentoring the next generation of engineers, I focus on scalable, high-uptime solutions. I'm always looking to connect with others working at the intersection of Full Stack and AI to share architectural insights and explore how we can push the boundaries of agentic workflows.
Networking
What I can offer
- ›Architectural expertise in deploying scalable AI systems
- ›Full-stack engineering mentorship
- ›Educational resources for AI/ML and FAANG interview prep
Looking for
- ›expanding my professional network
- ›exploring mutual opportunities in the intersection of Full Stack Engineering and AI Architecture
Best fit for
Current Interests
Background
Career
Began as an app developer and instructor, transitioned through full-stack roles at ING and HCL, specialized in API automation at Capital One, and currently architects AI advisory engines.
Education
M.S. in Business Analytics, University of New Haven (2022–2023); B.Tech in Computer Science, Sathyabama University (2016–2020)
Achievements
- ›Built 20+ applications serving over 1 million users with 99.99% uptime
- ›Reduced infrastructure costs by 30% via architectural refactoring
- ›Reduced graph traversal latency by 60% using Neo4j
- ›Educated 40,000+ students on Udemy
- ›Reduced manual effort by 80% through automated banking disclosures at Capital One
Opinions
- Theory is great, but building is better.
- The 'right' metric depends on the business problem; simple models often outperform complex ones on small datasets.
- Portfolios should show 'the grind' (GitHub streaks, LeetCode progress) rather than just highlights.