ECE · University of Washington · Class of 2027

Triaksha Singh

Systems engineer building ML, real-time systems, and trading infrastructure.

Electrical & Computer Engineering — University of Washington

Projects →Experience →Tech Stack →

GitHub Resume

Available · Summer 2026

03 /

Projects

Built from first principles — benchmarked, not just described.

Quant / Trading Systems

Limit Order Book & Matching Engine

Price-time priority limit order book with deterministic matching for market and limit orders, partial fills, and constant-time order lookup by ID.

▸O(log P) price-level updates and O(1) cancel via hash index + intrusive linked lists.
▸Benchmarked 1,000,000 synthetic events — 102,056 events/sec; p50 5.29 μs, p95 13.18 μs, p99 20.33 μs (Python).
▸Modular architecture separating order book state, matching logic, and benchmark harness.

102K evt/s

Throughput

5.29 μs

Latency p50

20.33 μs

Latency p99

PythonMarket MicrostructureOrder FlowData StructuresBenchmarking

GitHub

Quant / Risk Systems

Volatility Estimator Toolkit

Modular financial analytics engine for large-scale time-series processing with rolling-window and exponentially weighted statistical estimators.

▸Implemented EWMA and rolling-window estimators with configurable parameters — validated across varying input regimes.
▸Optimized high-volume time-series computation using vectorized NumPy operations, eliminating iterative loops.
▸Benchmarked estimator stability and computational trade-offs under simulated market shocks to evaluate robustness.

PythonNumPypandasstatsmodelsEWMATime SeriesBenchmarking

GitHub

ML / Healthcare

Diabetes Prediction Model

End-to-end machine learning pipeline for predicting diabetes using the Pima Indians Diabetes dataset — from preprocessing to single-user inference.

▸Standardized medical feature inputs using StandardScaler and applied stratified 80/20 train-test split to maintain class balance.
▸Implemented linear SVM classifier; evaluated model performance using accuracy metrics on both training and test sets.
▸Supports single-user predictions via manual feature input through a clean inference interface.

Pythonscikit-learnNumPypandasSVMStandardScaler

GitHub

02 /

Experience

Research Intern

Research

Allbritton Lab – UW Bioengineering

Feb 2026 — Present

Seattle, WA

▸Architected end-to-end communication pipeline between ESP32 firmware and Raspberry Pi controller using WiFi + structured JSON messaging, enabling reliable real-time device-to-server synchronization.
▸Designed and implemented distributed device state management logic across firmware, backend agents, and web dashboard — consistent execution of multi-stage workflows under concurrent user access.
▸Debugged and integrated multi-layer system components (embedded firmware, Linux-based controller, Node.js backend, MongoDB, Socket.IO telemetry) to centralize communication flows.

ESP32Raspberry PiNode.jsMongoDBSocket.IOLinuxWiFiJSON

AI/ML Intern

Internship

Inntech

Jun 2025 — Sep 2025

Muscat, Oman

▸Engineered modular Python data-processing pipelines for structured datasets, reducing manual overhead by ~40% while improving computational reliability and system maintainability.
▸Designed and validated predictive computation workflows with rigorous error analysis and cross-validation, integrating outputs into backend systems with clean, scalable architecture.

PythonScikit-learnPandasNumPystatsmodelsData Pipelines

Mechatronics Engineer

Research

UW Husky Robotics Team

Oct 2024 — May 2025

Seattle, WA

▸Built performance-critical distributed ROS2 services in C++/Python on Linux, reducing end-to-end control latency by ~30% through asynchronous messaging and execution profiling.
▸Integrated Betaflight flight-control stack to establish deterministic servo signal pipelines between transmitter input and actuator output.
▸Debugged cross-layer failures across embedded firmware, Linux services, and hardware interfaces to ensure stable real-time system behavior.

ROS2C++PythonLinuxBetaflightEmbedded FirmwareProfiling

04 /

Tech Stack

Languages

Python

C++

Java

HTML

CSS

JavaScript

MySQL

ML & Data

PyTorch

NumPy

pandas

scikit-learn

Infrastructure

Docker

Linux

Git

PostgreSQL

Systems

ROS2

CUDA

Node.js

MongoDB

05 /

Contact

Let's work together.

Open to SWE internships, quant research, and ML engineering roles for Summer 2026. Based in Seattle, WA.

Phone

+1 (206) 605-6394

GitHub

triakshasingh

Download Resume Send Email

triakshasingh@gmail.com · tsingh05@uw.edu · +1 (206) 605-6394

ECE · University of Washington · Class of 2027

Triaksha Singh

Systems engineer building ML, real-time systems, and trading infrastructure.

Electrical & Computer Engineering — University of Washington

Projects →Experience →Tech Stack →

GitHub Resume

Available · Summer 2026

03 /

Projects

Built from first principles — benchmarked, not just described.

Quant / Trading Systems

Limit Order Book & Matching Engine

Price-time priority limit order book with deterministic matching for market and limit orders, partial fills, and constant-time order lookup by ID.

▸O(log P) price-level updates and O(1) cancel via hash index + intrusive linked lists.
▸Benchmarked 1,000,000 synthetic events — 102,056 events/sec; p50 5.29 μs, p95 13.18 μs, p99 20.33 μs (Python).
▸Modular architecture separating order book state, matching logic, and benchmark harness.

102K evt/s

Throughput

5.29 μs

Latency p50

20.33 μs

Latency p99

PythonMarket MicrostructureOrder FlowData StructuresBenchmarking

GitHub

Quant / Risk Systems

Volatility Estimator Toolkit

Modular financial analytics engine for large-scale time-series processing with rolling-window and exponentially weighted statistical estimators.

▸Implemented EWMA and rolling-window estimators with configurable parameters — validated across varying input regimes.
▸Optimized high-volume time-series computation using vectorized NumPy operations, eliminating iterative loops.
▸Benchmarked estimator stability and computational trade-offs under simulated market shocks to evaluate robustness.

PythonNumPypandasstatsmodelsEWMATime SeriesBenchmarking

GitHub

ML / Healthcare

Diabetes Prediction Model

End-to-end machine learning pipeline for predicting diabetes using the Pima Indians Diabetes dataset — from preprocessing to single-user inference.

▸Standardized medical feature inputs using StandardScaler and applied stratified 80/20 train-test split to maintain class balance.
▸Implemented linear SVM classifier; evaluated model performance using accuracy metrics on both training and test sets.
▸Supports single-user predictions via manual feature input through a clean inference interface.

Pythonscikit-learnNumPypandasSVMStandardScaler

GitHub

02 /

Experience

Research Intern

Research

Allbritton Lab – UW Bioengineering

Feb 2026 — Present

Seattle, WA

▸Architected end-to-end communication pipeline between ESP32 firmware and Raspberry Pi controller using WiFi + structured JSON messaging, enabling reliable real-time device-to-server synchronization.
▸Designed and implemented distributed device state management logic across firmware, backend agents, and web dashboard — consistent execution of multi-stage workflows under concurrent user access.
▸Debugged and integrated multi-layer system components (embedded firmware, Linux-based controller, Node.js backend, MongoDB, Socket.IO telemetry) to centralize communication flows.

ESP32Raspberry PiNode.jsMongoDBSocket.IOLinuxWiFiJSON

AI/ML Intern

Internship

Inntech

Jun 2025 — Sep 2025

Muscat, Oman

▸Engineered modular Python data-processing pipelines for structured datasets, reducing manual overhead by ~40% while improving computational reliability and system maintainability.
▸Designed and validated predictive computation workflows with rigorous error analysis and cross-validation, integrating outputs into backend systems with clean, scalable architecture.

PythonScikit-learnPandasNumPystatsmodelsData Pipelines

Mechatronics Engineer

Research

UW Husky Robotics Team

Oct 2024 — May 2025

Seattle, WA

▸Built performance-critical distributed ROS2 services in C++/Python on Linux, reducing end-to-end control latency by ~30% through asynchronous messaging and execution profiling.
▸Integrated Betaflight flight-control stack to establish deterministic servo signal pipelines between transmitter input and actuator output.
▸Debugged cross-layer failures across embedded firmware, Linux services, and hardware interfaces to ensure stable real-time system behavior.

ROS2C++PythonLinuxBetaflightEmbedded FirmwareProfiling

04 /

Tech Stack

Languages

Python

C++

Java

HTML

CSS

JavaScript

MySQL

ML & Data

PyTorch

NumPy

pandas

scikit-learn

Infrastructure

Docker

Linux

Git

PostgreSQL

Systems

ROS2

CUDA

Node.js

MongoDB

05 /

Contact

Let's work together.

Open to SWE internships, quant research, and ML engineering roles for Summer 2026. Based in Seattle, WA.

Phone

+1 (206) 605-6394

GitHub

triakshasingh

Download Resume Send Email

triakshasingh@gmail.com · tsingh05@uw.edu · +1 (206) 605-6394