∇f(x)=0
[ 1 0 0 ]
[ 0 λ 0 ]
[ 0 0 μ ]
[ 0 λ 0 ]
[ 0 0 μ ]
∫₀^∞ e⁻ˣ dx = 1
det(A-λI)=0
P(A|B) = P(B|A)P(A)/P(B)
∂u/∂t = α∇²u
Ax = b
[ a b ] [ e f ]
[ c d ] × [ g h ]
[ c d ] × [ g h ]
ζ(s) = Σ 1/nˢ
O(n log n)
e^(iπ) + 1 = 0
∮ F·dr = 0
Σₙ aₙxⁿ
ker(T) ⊕ im(T)
∇²ψ + k²ψ = 0
φ = (1+√5)/2
σ² = E[(X-μ)²]
n! ~ √(2πn)(n/e)ⁿ
∑ 1/n² = π²/6
ℤ/nℤ
gcd(a,b)·lcm(a,b)=ab
∂f/∂x = ∂f/∂y
Hom(π₁, G)
dx ∧ dy
∫∫ K dA = 2πχ
Rᵢⱼₖₗ = gᵢₘRᵐⱼₖₗ
χ(S²) = 2
dim(V) = n
Engineer | Mathematician | Builder
I've worked on hard problems across domains-distributed systems, quantitative finance, game development, robotics, physics simulation. The common thread: math, engineering, and building from scratch or optimizing.
Im looking for hard problems. Got one? Let's Talk.
github: @Notuc
email: gale.nathann@gmail.com
Latest from the blog
BTC Predictions Using Deep Learning Models2023-10-20
Read post →The primary goal of this project is to develop a deep learning model that can accurately predict the future price movements of Bitcoin.
Shardkvopen source
Embedded Linux
Embedded Linux telemetry system built with both Buildroot and Yocto, running on Raspberry Pi 4. The same application stack is built twice — once with each build system — enabling a direct comparison of toolchains, boot times, image sizes, etc.
RTOS-PriorityInversion-WDopen source
Multi-Task FreeRTOS application
FreeRTOS application on the STM32F446RE in C/C++/CMake: five preemptive tasks, DMA-driven I2C sensor acquisition, mutex shared bus access with priority inheritance, multi-depth queue producer-consumer pipelines
Shardkvopen source
Distributed fault-tolerant KV store
Multi-threaded Raft consensus with election monitoring, leader heartbeats, log replication with conflict detection, and binary serialization using std::variant.
Match Engineopen source
GIF
Low-latency order matching
Lock-free SPSC ring buffers, price-time priority matching, ZeroMQ market data pipeline with snapshot recovery and L2 state printing.
Time Seires & Nueral Networksclosed source
Econometrics & neural networks
ARIMA models and LSTM networks for stock price forecasting. ADF/KPSS stationarity tests, VaR backtesting, achieving 96% R².
Code available on request.
Problems on Honeycomb Topologiesresearch
Domination & Independence on cylindrical honeycomb boards
Graph theory research on domination & independence numbers for cylindrical topologies. Monte Carlo algorithms, combinatorics proofs. Presented at MAA conference.
Full Paper/Code available on request.
BTC Deeplearning Predictionsclosed source
Statistical Learning
Monte Carlo option pricing, time-series forecasts, risk assessment for interest rates and credit spreads.
Details/Code available on request.
Heart Diseases Classificationclosed source
Classification Predictions
Monte Carlo option pricing, time-series forecasts, risk assessment for interest rates and credit spreads.
Details/Code available on request.
Quant-Finance Modelsclosed source
Quantitative finance research
Monte Carlo option pricing, Time-series forecasts, Finite Difference, Numerical Methods, Risk assessment for interest rates and credit spreads.
Details/Code available on request.
+
More projects on GitHub