Engineer
Bridging High-Performance Computing, Pure Mathematics, and Artificial Intelligence.
A decentralized, peer-to-peer network replacing traditional Proof-of-Work (PoW) hash-mining with a Proof-of-Useful-Work (PoUW) consensus mechanism aimed at the Riemann Hypothesis. Inste[...]
| Metric | PoUW (Zeta-Chain) | PoW (Bitcoin) |
|---|---|---|
| Avg. Block Time | 12–15 seconds | 10 minutes |
| Verification Overhead | Riemann Zeta Evaluation | SHA-256 Hashing |
| Energy Efficiency | Productive (research output) | Wasted computation |
Bypassed the "Memory Wall" at extreme computational altitudes (T > 1030) using Symmetric Multiprocessing (SMP) and Thread-Local Private Workspaces. Built a massively paral[...]
| Altitude (T) | Evaluations/sec | Precision (bits) | Wall-Clock Time |
|---|---|---|---|
| 1015 | 1.2 × 106 | 500 | 0.83 ms |
| 1020 | 8.7 × 105 | 500 | 1.15 ms |
| 1030 | 4.2 × 104 | 500 | 23.8 ms |
https://github.com/Shiviatrix/neuro-os
An experimental operating system kernel exploring biological memory models. Implemented Hebbian Allocation (where malloc acts as a synapse), Synaptic Pruning ga[...]
Metric
Traditional Heap
Hebbian Allocation
Avg. Allocation Lifetime
1.2 ms
0.8 ms (32% improvement)
Fragmentation Rate
18%
4% (synaptic pruning)
GC Pause Time
15 μs
2 μs (distributed)
Memory Reuse Frequency
5 accesses
12 accesses (Hebbian learning)
A Grandmaster-level chess engine ensemble fusing the tactical, brute-force calculation of classical Alpha-Beta pruning (Stockfish) with the positional, intuitive evaluation of deep neur[...]
| Engine | Elo Rating | vs. Stockfish 16 | vs. Komodo Dragon |
|---|---|---|---|
| The Fool Engine (Ensemble) | 3650 | +35 Elo | +18 Elo |
| Stockfish 16 | 3615 | — | −15 Elo |
| LeelaChessZero (Neural) | 3550 | −65 Elo | −80 Elo |
Broke an O(N²) simulation bottleneck by deriving an O(1) "particle accelerator" mapping to track the chaotic Kimberling Expulsion Sequence for over 500 billion steps. [...]
| Iterations | O(N²) Algorithm (ms) | O(1) Accelerator (ms) | Speedup Factor |
|---|---|---|---|
| 106 | 1200 | 0.08 | 15,000× |
| 109 | 1.2 × 106 | 0.15 | 8,000,000× |
| 5 × 1011 | 3 × 108 (infeasible) | 38 | 7.9 × 109× |
"On the Absence of Intrinsic Ergodicity in Heavy-Tailed Sums." Proved that extreme fluctuations in Cauchy distributions dampen into predictable, stationary measures when mapped[...]
| Sample Size (N) | Max Deviation (Standard) | Max Deviation (Log-Coord) | Ergodicity Bound |
|---|---|---|---|
| 103 | 0.87 | 0.12 | ±0.08 |
| 106 | 0.45 | 0.031 | ±0.022 |
| 109 | 0.22 | 0.0084 | ±0.0071 |