Jake Vichnis brings a unique blend of creative vision and analytical rigor to the worlds of technology, quantitative finance, and AI. Though his roots lie in music production—shaped by studio work on Grammy-winning projects at Progressive Musik Group and years guiding aspiring artists at Third Street Music School Settlement—this site highlights his evolution from artistic creator to quant developer and AI specialist.
Jake began as a library technician while studying Audio Technology at The City College of New York before honing his technical skills as a production assistant on Grammy-winning projects. Drawing on those studio experiences, he began teaching and designed a hands-on music production curriculum for students-teaching them how to craft beats, mix tracks, and tell stories through sound. That formative chapter instilled in him a deep appreciation for pattern, structure, and disciplined practice.
Inspired by the patterns of music, Jake wrote his first Python script, quickly recognizing that algorithms—like melodies—could solve real-world problems. What began as simple data parsing soon ignited a passion for the mathematical foundations beneath every computation.
From 2022 to 2025 Jake built the proper mathematical foundation and immersed himself in advanced quantitative methods: deploying multivariate calculus to fine-tune gradient descent routines, applying matrix decompositions and eigenvalue analysis to build and invert covariance matrices for portfolio optimization, and leveraging probability theory and statistical inference to model risk and forecast market behavior. He put theory into practice with projects like his Gradient Descent Portfolio Optimizer, where custom partial-derivative implementations and adaptive line searches drive asset-weight calculations, and by architecting an upcoming HFT Strategy Engine that relies on discrete-math–informed data structures and sub-microsecond execution in C++.
On the AI front, Jake’s C++–based multiple linear regression program laid the groundwork for Python experimentation. He explored tensor operations and automatic differentiation in PyTorch—applying concepts from linear algebra (tensor contraction, singular value decomposition) and optimization theory (momentum methods, adaptive learning rates)—to prototype neural networks and improve convergence on classification tasks.
Now completing his B.S. in Computer Science at Western Governors University (graduating October 2025), Jake is laser-focused on roles and internships that merge:
1. Quantitative Development -
Crafting algorithmic trading systems in C++ for minimal latency
2. Data Science & AI/ML –
Building scalable models and pipelines with PyTorch and TensorFlow
3. Cloud & DevOps –
Containerizing microservices with Docker and deploying to cloud platforms
He’s actively seeking opportunities in tech-driven finance, where his creative intuition, mathematical inclination, and full-stack coding expertise can deliver real-world impact.