>yusuf.sys ~ init

Yusuf
AI Systems Engineer

PythonTypeScriptAWSPyTorchOptimization
>yusuf.sys ~ about

The Engineer Behind the Systems

Started as a software engineer, building things that work. Shipped production code, learned how systems break, and developed an obsession with reliability.

Moved into machine learning and data science, building things that learn. Then into data engineering, architecting the pipelines that feed intelligence. Each transition added a new layer to how I think about systems.

Now at the intersection: AI Systems Engineering. I don’t just plug in models, I design the full system around them. With a foundation in operations research, I build systems that don’t just predict, they optimize and decide.

SE
Software EngineeringFoundation

Where it started

ML
Machine Learning & Data ScienceGrowth

Teaching machines to see patterns

DE
Data EngineeringExpansion

Architecting the pipelines behind intelligence

AI
AI Systems + Operations ResearchNow

Where I live now — optimization meets intelligence

>yusuf.sys ~ capabilities

Three things I do that most engineers treat as separate disciplines

Intelligent Systems

I integrate AI models into production systems, not as demos, but as decision-making engines embedded in real workflows.

  • >Claude API, Gemini, HuggingFace model integration
  • >Agentic workflows and orchestration
  • >AI-powered automation pipelines
  • >Real-time inference systems
Claude APIGeminiLangChainPython

Optimization & Decision Models

With a background in operations research, I build mathematical models that don’t just predict, they prescribe the best action.

  • >Linear and integer programming
  • >Descriptive, predictive, prescriptive analytics
  • >Simulation and scenario modeling
  • >Multi-objective optimization
OR-ToolsPuLPPyTorchSciPy

End-to-End Engineering

From raw data to deployed model to cloud infrastructure, I can own the full stack without handing off to three different teams.

  • >Data pipelines and feature engineering
  • >Model training, evaluation, deployment
  • >Cloud infrastructure (AWS)
  • >Full-stack application development
AWSNext.jsTypeScriptdbt
>yusuf.sys ~ projects

Systems I've Built

End-to-end. From model to deployment.

In DevelopmentResearch Project

NETI–HyOptima

Net-Zero Energy Transition Intelligence

A cloud-native AI decision intelligence platform that translates energy transition policies into optimized, data-driven strategies for hybrid energy systems spanning gas, renewables, storage, and hydrogen.

  • MILP optimization engine (Pyomo) minimizing cost, emissions, and unserved energy
  • ML forecasting stack: LSTM + XGBoost + Prophet for demand and renewable prediction
  • Monte Carlo stochastic simulation for uncertainty in demand, fuel prices, weather
PythonPyomoLSTMXGBoostProphet+4
CompleteData Engineering

Real-Time Energy Data Pipeline

End-to-End Cloud-Native Data Engineering

A portfolio-grade cloud-native pipeline that simulates, streams, processes, stores, transforms, optimizes, and visualizes energy distribution data — with a two-stage Operations Research optimizer for fair energy allocation.

  • Real-time IoT simulation → Kafka → Spark → PostgreSQL + S3 pipeline
  • Stage 1 LP optimizer: fair energy allocation across 5 zones by priority
  • Stage 2 Transportation optimizer: routes power minimizing transmission loss
KafkaApache SparkdbtPuLPAirflow+5
LiveFull-Stack AI

HabitOS

AI-Driven Behavioral Optimization Platform

A production-grade full-stack decision support system that transforms life goals into mathematically optimized daily schedules using Mixed-Integer Linear Programming.

  • MILP solver (PuLP + CBC) optimizing daily schedule across time, energy, and behavioral constraints
  • Solves typical schedules in under 500ms — near real-time performance
  • JWT authentication, role-based access, async FastAPI backend
FastAPIReact 18TypeScriptPuLPPostgreSQL+4
CompleteML + Operations Research

Titanic Survival & Lifeboat Optimizer

ML Prediction + Operations Research Allocation

A full-stack decision intelligence system fusing XGBoost survival prediction with Mixed-Integer Programming to solve the lifeboat allocation problem under ethical and capacity constraints.

XGBoostPuLPFastAPIStreamlit+4
CompleteAI Integration

ExpenseWise

AI-Powered Expense Analysis Platform

An AI-integrated expense analysis service using OpenAI and Google Gemini for personalized financial forecasting and smart budgeting recommendations.

FlaskOpenAIGemini APIPostgreSQL+1
>yusuf.sys ~ stack

Tools I Build With

AI & Intelligence

PythonPyTorchXGBoostscikit-learnHuggingFaceMLflowPuLPPyomoOR-Tools

Data Engineering

Apache KafkaApache SparkApache AirflowdbtPostgreSQLBigQueryDuckDBAWS S3Parquet

Backend & APIs

FastAPIFlaskNext.jsTypeScriptRedisSQLAlchemyJWTOpenAPI

Cloud & Infrastructure

AWS (S3, EC2, Lambda, SES, DynamoDB)GCPTerraformDockerDocker ComposeNginxGitHub Actions

Frontend & Visualization

React 18Tailwind CSSFramer MotionThree.jsStreamlitPlotlyRecharts
>yusuf.sys ~ contact

Let's Build Something

Open to roles, collaborations, and interesting problems.

Open to Opportunities