Open to opportunities

Muhammad Zainurrahman

@muhammadzainurrahman

Message

AI Engineer specializing in production-grade autonomous multi-agent systems and LLM orchestration pipelines.

Indonesia

Message

What I'm looking for

I’m looking to build production-grade autonomous multi-agent LLM systems—shipping FastAPI/Docker services on Google Cloud Run. I want teams that value rigorous data integrity, scalable MLOps, and real-time impact through vision-language and API orchestration.

I’m a results-driven AI Engineer focused on designing and deploying production-grade autonomous multi-agent systems and LLM orchestration pipelines. I architect agentic workflows with DeepSeek, Gemini, Mistral, and OpenRouter, and I ship end-to-end cloud deployments on Google Cloud Run using FastAPI and Docker.

I currently work as an independent AI engineer and researcher, building a production-grade Multi-Agent Financial Intelligence System that analyzes real-time Stocks, Crypto, and Commodities with zero human intervention. I’ve created a hybrid Reasoning + Vision agent backend that can analyze stock charts and synthesize geopolitical news from live web sources, and I’ve deployed a Stock Forecasting platform using NeuralForecast (N-BEATS) with an agentic reasoning layer across 4 Docker microservices on Google Cloud Run.

I also build ML systems with a strong reliability mindset: I engineered a customer churn prediction pipeline using PyCaret and XGBoost with validated 88% Accuracy and 0.91 AUC, and refactored the feature pipeline after identifying temporal leakage. For booking intent, I designed an end-to-end pipeline that addresses an 85% drop-off class imbalance using SMOTE within strict cross-validation, then deployed the final model as production-grade FastAPI/Docker microservice with a React (Vite) dashboard and a Pull-Based GitOps CI/CD flow.

Experience

Work history, roles, and key accomplishments

Current

Independent AI Engineer

Current

Self-Employed / Portfolio

Jan 2024 - Present (2 years 6 months)

Architected a production-grade multi-agent financial intelligence system for analyzing real-time stocks, crypto, and commodities with zero human intervention. Built hybrid reasoning and vision agent backends, deployed FastAPI services in Docker on Google Cloud Run, and delivered real-time Telegram notifications.

Deepseek R1 AsyncIO Multi Agent Systems OpenRouter Mistral Vision Language Models fastAPI Docker Google Cloud Run NeuralForecast NBEATS Microservices Architecture Context Aware Multi Modal Pipelines

Data Science Job Simulation

Lloyds Banking Group

Nov 2025 - Dec 2025 (1 month)

Engineered a customer churn prediction system using PyCaret and XGBoost, achieving 88% accuracy and 0.91 AUC after resolving data leakage issues. Refactored and deployed the ML workflow as a FastAPI REST API with Pydantic validation and built a React dashboard for real-time risk visualization.

PyCaret XGBoost Feature Engineering Temporal Leakage Detection fastAPI Pydantic REST APIs Docker Model Deserialization Hot Reloading

Data Science Job Simulation

British Airways

Dec 2025 - Present (7 months)

Built an end-to-end ML pipeline to predict customer booking intent, addressing severe class imbalance (85% drop-off) with SMOTE in a leakage-safe cross-validation setup. Tuned a Random Forest model (86% accuracy, AUC-ROC 0.82) and deployed it as a FastAPI + Docker microservice with a React (Vite) frontend and GitOps CI/CD model fetching from a GCP MLflow tracking server.

Booking Intent Prediction Class Imbalance Handling SMOTE Cross Validation Random Forest & Model Tuning Chi Square Statistical Tests fastAPI Docker Vite Microservices Architecture GitOps MLFlow Model Registry Champion Model Fetching

Education

Degrees, certifications, and relevant coursework

Brawijaya University

Bachelor's degree in Physics, Physics

2018 - 2022

Grade: GPA: 3.72

Bachelor's degree in Physics at Brawijaya University (2018–2022). Thesis: Optimization of Flattening Filter Energy for Geiger Muller Detectors Using MCNPX Simulation (GPA: 3.72).