01
Experimental Deep Learning
Primary discipline. Building and breaking models in PyTorch for its flexibility — favoring local training on Apple Silicon and lightweight LLMs tuned for the M1 Pro.
// Informatics Eng. · UKSW · Batch 2024
N 7°19′ · Salatiga, ID
Yohanes Widi Prasetyanto
AI practitioner working in PyTorch, Web3 explorer shipping Solidity & Rust, and a speculative quant translating market noise into signal. I build logic that is complex yet intelligible.
Deep Learning
Blockchain / Web3
Quant Trading
Teaching Assistant
01
Domains of focus
02
Blockchain & Web3
Advanced beginner scaling fast. Writing smart contracts in Solidity on Hardhat, exploring Rust, and deploying through MetaMask toward production-grade on-chain systems.
03
Quantitative Speculation
High-volatility assets read through indicator patterns. Transitioning from pure speculation toward data-science-driven technical & fundamental analysis.
02
Operating thesis
“Data is the ultimate gold; those who command data, command the world.” — core philosophy
A selective networker with high standards for the rooms I enter — direct at first, deeply collaborative once aligned. I value people who are smarter, sharper, or carry a higher vision.
As a teaching assistant I run a roaming, hands-on approach: fast, accessible, and obsessed with developing student logic over stale theory. In competitions and organizations I switch to strict self-discipline and hyper-efficient execution.
03
Technical DNA
AI / ML
- PyTorch core
- Python native
- Deep Learning focus
- Reinforcement L. exp
- Local LLMs M1
Web3
- Solidity contracts
- Hardhat framework
- Rust growing
- MetaMask deploy
Environment
- Google Colab heavy
- VS Code daily
- Antigravity IDE +codex
- MacBook M1 Pro 16GB
Markets
- Crypto Assets active
- Penny Equities vol
- Indicator Reads ta
- Quant Bridge next
04
Roles in play
2024 →
Pedagogy
Teaching Assistant (Asdos)
Hands-on, real-case instruction focused on logical structuring and student reasoning.
2024 →
Leadership
Operation Crew
Selective leadership under high-stakes timelines with disciplined, efficient execution.
Active
Competition
Competitive Event Competitor
Performance mode: strict self-discipline, hyper-efficient time management, results first.
Ongoing
Academic
Informatics Engineering · UKSW
Batch 2024, Semester 4 — building the technical foundation for an AI-defined future.
// 05 — Open the channel
Let's build something
Let's build something
that commands data.
I keep a tight circle, but I'm always open to people building toward a higher vision. Bring the signal — skip the noise.
Initiate contact →