0%
About Vision Impact Services Experience Projects Skills Contact
Muhammad Ali Mahmood
🎓
Erasmus Mundus
Scholar
💼
3+ Years
Experience
About Me

Engineering Excellence
From Backend to AI

I'm Muhammad Ali Mahmood, a software engineer who thrives at the intersection of robust backend systems and intelligent AI solutions. Based in the Netherlands, I'm completing my Master's in AI under the prestigious Erasmus Mundus scholarship.

My journey has spanned continents—from building enterprise microservices for Fortune 500 clients at GoSaaS AI, to developing vision-language models at ASML, to conducting research across labs in Pakistan, Spain, Slovenia, and the Netherlands.

I don't just write code—I architect solutions. Whether it's designing Spring Boot microservices, building RAG pipelines for knowledge retrieval, or fine-tuning LLMs for domain-specific tasks, I bring engineering rigor and AI innovation together.

0
Years Experience
0
Projects Delivered
0
Countries Worked
Scroll to explore
Open to opportunities — Based in Netherlands

Crafting Intelligent Systems at Scale

Backend Engineer & AI Practitioner with 3+ years building production systems, LLM applications, and scalable microservices. Currently pursuing MSc in AI at Radboud University as an Erasmus Mundus Scholar.

Explore My Work Get In Touch
Results That Matter

Measurable Impact

Tangible value delivered to clients and organizations worldwide

🏢
0
Enterprise Clients Served

Delivered production systems for Fortune 500 companies and industry leaders

Intel Roche PureStorage Renesas
🚀
0
Production Microservices

Deployed scalable Spring Boot services with auth, monitoring, and CI/CD pipelines

🌍
0
Companies Collaborated

Worked across USA, South Korea, Netherlands, and remote teams globally

🔬
0
Data Points Processed

Built pipelines handling millions of records for analytics and ML applications

What I Do

Services & Expertise

Three specialized domains with deep technical proficiency

⚙️

Full-Stack Development

End-to-End Solutions

Building complete applications from database to UI. Specializing in scalable backend systems with Java Spring Boot, REST APIs, and microservice architectures, complemented by modern React frontends. Full ownership from design through production deployment.

Java Spring Boot REST APIs Microservices ReactJS PostgreSQL Docker CI/CD AWS
🧠

LLM Engineering

Large Language Model Solutions

End-to-end LLM application development—from building RAG pipelines and semantic search to fine-tuning models for specific domains. Deep understanding of transformer internals, prompt engineering, and deploying LLM systems in production with proper MLOps practices.

RAG Pipelines LangChain Fine-tuning Prompt Engineering Vector DBs Hugging Face NLP Information Retrieval
🤖

AI Engineering

Production ML Systems

Taking AI from research to production. Model training, evaluation, and deployment with proper MLOps infrastructure. Computer vision systems, VLMs, and multimodal AI. Deep expertise in PyTorch, experiment tracking, and scalable inference pipelines.

PyTorch TensorFlow MLflow MLOps Computer Vision VLMs Model Deployment YOLO
Career Journey

Professional Experience

AI/ML Intern

ASML Holdings • Netherlands
Feb 2026 – Present
  • Building multimodal benchmarks for semiconductor diagnostics using vision-language models
  • Fine-tuning VLMs/LLMs on internal cases to improve fault identification accuracy
  • Developing reproducible evaluation pipelines with Python, PyTorch, Hugging Face, and MLflow

Software Engineer

GoSaaS AI • Remote
June 2023 – July 2025
  • Delivered 10+ Java Spring Boot microservices and REST APIs for Intel, Roche, and PureStorage
  • Improved data synchronization by ~30% through backend optimization on Oracle Cloud & Agile PLM
  • Containerized and deployed model-serving APIs with authentication, routing, and monitoring
  • Built reporting services with Spring Data JPA, integrated CI/CD pipelines with Docker and Jenkins

Computer Vision Research Assistant

MachVIS Lab, NUST • Pakistan
June 2022 – June 2023
  • Curated dataset of 3,000+ radiometrically calibrated multispectral UAV images
  • Designed YOLOv7-based detection model achieving 93% mAP, reducing manual analysis by 50%

Software Engineer Intern

SKAI Worldwide • Seoul, South Korea
Jan 2023 – June 2023
  • Built fault-tolerant Oracle→PostgreSQL migration tool with checkpointing and resumable execution
  • Implemented multi-threaded workers with batching, structured logging, and CLI configuration

Software Engineer Intern

CONFIZ Ltd. • Bellevue, USA
July 2021 – Sep 2021
  • Implemented Spring Boot backend for employee offboarding with REST APIs, RBAC, and SQL Server
  • Integrated services with ReactJS frontend and HR systems for consistent data flow
  • Recognized as Best Performing Summer Intern
Featured Work

Selected Projects

A showcase of impactful work across AI, backend, and full-stack development

Research • Master's Thesis • ASML

VLM Benchmark for Semiconductor Diagnostics

Designing domain-specific benchmarks for ASML diagnostic workflows with multimodal test cases (logs, plots, screenshots) and a semi-automated annotation and evaluation suite.

Python PyTorch Hugging Face MLflow
Confidential
📄
📚
🔍
📊
🧠
Research • NLP • RAG

Uncertainty Estimation for Long-Form RAG

Built a RAG pipeline with BM25 and cross-encoder reranking over 6M+ Wikipedia passages, implementing MARS, Eccentricity, and novel SEEW uncertainty scoring.

Qwen2.5-7B SAFE BM25 FAISS
Repository
{"api": "v1"}
GET /users
POST /data
import React
export default
const App
@RestController
@GetMapping
return json
SELECT * FROM
JOIN tables
WHERE id
async fetch
await response
.then(data)
docker build
kubectl apply
npm start
REST → React UI
Hackathon • Full-Stack • LLM

Legacy UI Facade Maker

LLM-powered platform that inspects legacy REST/SOAP/OpenAPI services, infers normalized schemas, and auto-generates modern React CRUD dashboards without changing the backend.

Spring Boot LangChain React OpenAI
Repository Live Demo
Research • LLMs • Robustness

Improving Prompt Sensitivity in LLMs

Analyzed prompt sensitivity across 4 instruction-tuned LLMs using POSIX and embedding-based similarity, finding CoT and prompt voting improve stability more reliably than fine-tuning.

LLaMA Mistral LoRA Chain-of-Thought
Repository
Industry • Data Engineering

Transport Mode Inference from Mobile Data

Built a denoising pipeline for 900M cellular pings (→130M clean samples), revealing transport modes and commuting flows across Slovenian regions using clustering and spatial analysis.

Python DBSCAN K-Means SQL
Repository
Bachelor's Thesis • Computer Vision

Sunflower Yield Estimation System

UAV-based multispectral imaging pipeline with YOLOv7 detection achieving 93% accuracy for automated crop head counting. Includes React dashboard for field visualization.

YOLOv7 PyTorch OpenCV React
Repository Live Demo
Technical Arsenal

Skills & Technologies

💻

Languages

Java Python JavaScript TypeScript SQL
⚙️

Backend & Cloud

Spring Boot REST APIs Microservices Docker Kubernetes AWS Oracle Cloud Jenkins
🧠

LLMs & NLP

LangChain RAG Fine-tuning Hugging Face Transformers Vector DBs spaCy NLTK
🤖

AI & ML

PyTorch TensorFlow Scikit-learn MLflow MLOps YOLO OpenCV
🎨

Frontend & Tools

ReactJS Next.js Git Jupyter VS Code JIRA
🗄️

Databases

PostgreSQL Oracle MongoDB SQL Server Redis
Academic Background

Education

MSc in Artificial Intelligence

Radboud University Nijmegen

Netherlands • Spain • Slovenia

Sept 2024 – Aug 2026

  • Erasmus Mundus MSc in AI (EMAI) Full Scholarship
  • Rotated across 3 countries during 4-semester program
  • UPF (Spain), University of Ljubljana (Slovenia), Radboud (Netherlands)
  • Master's Thesis at ASML on Vision-Language Models

BS in Computer Science

National University of Sciences & Technology (NUST)

Islamabad, Pakistan

Sept 2019 – June 2023

  • CGPA: 3.76 / 4.00
  • 1st Prize for Best Industrial Final Year Project
  • 3× Dean's Honor List
  • Teaching Assistant for AI, DSA, and Algorithms
Get In Touch

Let's Build Something
Remarkable Together

Whether you have a challenging project, need AI/backend expertise, or just want to connect— I'd love to hear from you.

muhammad.ali.hawk@gmail.com