Available for opportunities · Munich, Germany

Mohd Zamin Quadri

|

AI Engineer crafting intelligent systems at the intersection of Mathematics, Machine Learning, and Deep Learning. Based in Munich, Germany.

0+ Years Experience
0+ Projects Delivered
0+ Certifications
0+ Roles Held
Scroll to explore

About Me

I'm an AI Engineer and Applied Mathematician based in Munich, Germany, currently completing my Master's in Mathematics in Science and Engineering at the Technical University of Munich.

My expertise spans the full ML lifecycle — from translating business problems into technical specifications, to building production-grade pipelines, to deploying uncertainty-aware models at scale. I've delivered solutions for clients in healthcare (AI Radiologist), insurance, automotive (Audi AG), and transportation policy.

I thrive at the intersection of rigorous mathematics and practical engineering, believing that the best AI systems are built on strong theoretical foundations and clean, reproducible code.

Location Munich, Germany
University TU Munich (MSc)
Current Role AI Engineer @ BP-ITCS
Languages English · German · Hindi
Specialization ML · GNN · Uncertainty QT
Thesis Topic GNN Surrogates + UQ

Professional Journey

Current

AI Engineer

BP-ITCS

July 2025 – Present Munich, Germany
  • Partnered with stakeholders to translate business needs into ML problem statements and KPIs
  • Designed end-to-end ML pipelines: data ingestion → feature engineering → training → evaluation
  • Delivered AI Radiologist prototype (PoC) — dataset prep, model experimentation, benchmarking
  • Developing ML solution for insurance client: feature design, model selection, calibration
  • Applied MLOps best practices: version control, experiment tracking, deployment-ready packaging
Python TensorFlow MLOps Computer Vision
Current

Master's Thesis Researcher

Technical University of Munich

August 2025 – Present Munich, Germany

Uncertainty Quantification for GNN Surrogates in Traffic Policy Modeling

  • Developing GNN surrogates for agent-based transport simulations in unseen cities
  • Designing uncertainty quantification methods with predictive confidence calibration
  • Investigating cross-scale prediction and multi-policy generalization
  • Exploring multi-fidelity learning for low/high-fidelity simulation integration
  • Benchmarking architectures (Transformer Convolutions) on accuracy and calibration
GNN PyTorch Geometric Uncertainty Quantification

Seminar Researcher

Technical University of Munich

October 2024 – March 2025

Identification of Neural Networks (Prof. Massimo Fornasier)

  • Formalized identifiability criteria for deep nets; mapped failure modes to debugging playbooks
  • Derived activation-dependent uniqueness conditions guaranteeing parameter identifiability
  • Operationalized Fefferman's framework into implementable checks
  • Characterized network isomorphisms and proposed symmetry-breaking strategies
Neural Network Theory Mathematical Analysis

Student Research Assistant

Technical University of Munich — Programming & Visualization

August 2023 – March 2024
  • Curated hands-on labs in Python, MATLAB, R, SQL for applied problem-solving
  • Conducted live-coding sessions with runnable examples
  • Standardized starter code and lab templates; mentored students on best practices
Python MATLAB R SQL Teaching

Intern — Workflow & Database Engineering

AUDI AG

January 2023 – June 2023 Ingolstadt, Germany
  • Engineered Excel-VBA workflows linking backend databases and standardizing reporting
  • Integrated multiple data sources into unified KPI views
  • Automated recurring reports, reducing data-management time ~50%
  • Implemented strict validation and error-handling for data accuracy
VBA SQL Excel Automation

Student Research Assistant — MATLAB Visualization

Technical University of Munich

April 2022 – December 2022
  • Architected MATLAB lab modules for numerical methods and visualization
  • Implemented vectorized numerical routines and benchmarked performance
  • Modeled systems using MATLAB/Simulink toolboxes with documented workflows
MATLAB Simulink Numerical Methods

Summer Research Intern

Indian Institute of Science Education and Research (IISER Bhopal)

May 2021 – July 2021 Bhopal, India
  • Surveyed Li-ion SOC/SOH literature; mapped methods to data/model assumptions
  • Implemented regression and clustering baselines on EV datasets
  • Assessed genetic-fuzzy subtractive clustering feasibility
ML Research Battery Analytics Clustering

Academic Foundation

Master of Science

Mathematics in Science and Engineering

Technical University of Munich

🇩🇪 Munich, Germany October 2025
Machine Learning Optimization Graph Neural Networks Uncertainty Quantification

Bachelor of Science (Hons.)

Mathematics

Aligarh Muslim University

🇮🇳 Aligarh, India March 2021
Pure Mathematics Applied Mathematics Statistics

AI Work Visualized

Real diagrams from my thesis, research, and project work

MLOps Pipeline · BP-ITCS
Live at Work

End-to-End ML Pipeline

Production ML workflow: from business requirements to deployed, monitored models.

📋 Business KPIs 🗄️ Data Ingestion ⚗️ Feature Engineering 🧠 Training + Calibration 🚀 Deploy + Monitor Hyperparameter Tuning Loop Requirements Pandas · SQL Scikit-learn TensorFlow · XGBoost MLflow · Docker
MLOpsPythonTensorFlowScikit-learn
Master's Thesis · TU Munich
Active Research

GNN Surrogate Architecture

Graph Neural Network replacing expensive transport simulations with uncertainty quantification.

City Graph Nodes=zones Edges=roads GCN Layer 1 ReLU + BN 128-dim GCN Layer 2 ReLU + BN 256-dim Transformer Conv Multi-Head Attention 512-dim μ (Mean) Prediction σ² (Var) Uncertainty Policy Output Trip demand Modal split ─────── 95% CI bands Calibrated UQ Message Passing Attention Dual Head
PyTorch GeometricGNNUncertainty QTTransformers
Seminar · Prof. Fornasier · TU Munich
Research

Neural Network Identifiability

Formalizing when neural network parameters can be uniquely recovered — symmetry breaking strategies.

Network A: θ₁ Neuron permutation symmetry Network B: θ₂ ≠ θ₁ f(x; θ₁) = f(x; θ₂) Same function! Key Result Activation-dependent uniqueness conditions → Fefferman framework → Symmetry breaking → Identifiability criteria (Prof. Fornasier, 2024)
Neural Network TheoryMathematical AnalysisIdentifiability
AI Radiologist PoC · BP-ITCS
Production PoC

Model Benchmarking

Comparative model evaluation across key metrics for the AI Radiologist prototype.

Accuracy Precision Recall F1 Score AUC-ROC
CNN Baseline
Accuracy88%
F1 Score84%
AUC-ROC0.90
ResNet-50
Accuracy92%
F1 Score90%
AUC-ROC0.95
Computer VisionMedical ImagingModel Evaluation

Featured Projects

🏥

AI Radiologist Prototype

End-to-end deep learning system for medical image analysis. Built dataset pipeline, experimented with CNN architectures, and delivered a demo-ready prototype for stakeholder review at BP-ITCS.

PythonTensorFlowOpenCVMedical Imaging
🚗

GNN Surrogates for Traffic Policy

Graph neural network models that replace expensive agent-based simulations to predict policy effects in unseen cities. Features uncertainty quantification and multi-fidelity learning.

PyTorch GeometricGNNUQTransformers
🛡️

Insurance Risk Prediction

ML solution for an insurance client featuring iterative model selection, calibration, thresholding, and performance improvements aligned with business KPIs and risk metrics.

PythonScikit-learnXGBoostMLOps
💬

NLP Sentiment Classifier

Fine-tuned BERT for multi-class sentiment analysis on product reviews. Implemented data augmentation, attention visualization, and achieved 94% F1-score with optimized inference pipeline.

HuggingFaceBERTNLPPyTorch
👁️

Real-Time Object Detection

Built a YOLOv8 detection pipeline for autonomous driving scenarios. Custom-trained on urban scenes with data augmentation, achieving real-time inference at 45 FPS on edge devices.

YOLOv8Computer VisionONNXTensorRT
📈

Energy Demand Forecasting

LSTM and Transformer-based time series model for predicting energy consumption patterns. Implemented multi-horizon forecasting with attention mechanism and temporal fusion transformers.

LSTMTransformersTime SeriesTensorFlow
🎯

E-Commerce Recommender System

Collaborative filtering and neural collaborative filtering approach for product recommendations. Implemented hybrid model combining content-based and interaction-based signals.

PythonNeural CFMatrix FactorizationFastAPI
📊

Visual Data Analytics Dashboards

Interactive Tableau dashboards with cross-filters and drill-downs. Visualized flow and volume fields in ParaView using streamlines, slices, and isosurfaces for scientific data.

TableauParaViewData Visualization
⚙️

Supply Chain Management Dashboards

Power BI dashboards for a car-parts manufacturer exposing critical metrics: time savings, ROI, defect rate, and perfect order rate. Earned stakeholder praise for intuitive UI/UX.

Power BISQLExcelSupply Chain
🔧

Car Parts Database Optimization

Migrated Audi's ad-hoc spreadsheet workflows to structured VBA modules. Built a macro-driven UI that reduced data-validation time by ~90% with unified ownership and approval views.

VBAExcelSQLAutomation
🧪

End-to-End ML Pipeline Suite

Orchestrated complete ML pipelines on real datasets covering supervised and unsupervised learning. Implemented SVM, classification, regression, and clustering with rigorous hyperparameter search.

Scikit-learnTensorFlowSVMClustering
🔍

Anomaly Detection in IoT Streams

Built an autoencoder-based anomaly detection system for IoT sensor data streams. Integrated statistical process control with deep learning for multi-modal anomaly scoring.

AutoencodersStreaming DataPythonKafka
🎨

GAN-Based Image Synthesis

Implemented StyleGAN2-ADA for high-resolution image generation with adaptive discriminator augmentation. Achieved FID score of 12.3 on custom dataset with progressive growing and style mixing.

StyleGAN2PyTorchGANsCUDA
🎲

Bayesian Hyperparameter Optimization

Built a framework for efficient model tuning using Gaussian Process-based Bayesian optimization. Reduced search time by 70% compared to grid search while finding better configurations.

BoTorchGPyTorchOptunaPython
🤖

LLM Retrieval-Augmented Generation

Designed a RAG pipeline using LangChain and vector databases for domain-specific Q&A. Implemented chunking strategies, embedding fine-tuning, and re-ranking for improved answer quality.

LangChainChromaDBOpenAIFastAPI
🕹️

Deep Reinforcement Learning Agent

Trained PPO and DQN agents for complex control tasks in simulation environments. Implemented reward shaping, curriculum learning, and multi-agent coordination strategies.

Stable-Baselines3GymnasiumPPOPyTorch
🔗

Multi-Modal Data Fusion

Built a cross-attention model fusing tabular, image, and text data for patient outcome prediction. Implemented late fusion and early fusion architectures with attention-weighted modality gating.

TransformersCross-AttentionTensorFlowMulti-Modal
🌐

Privacy-Preserving Federated Learning

Implemented federated averaging with differential privacy for decentralized model training across institutions. Achieved 96% of centralized accuracy with zero data sharing.

FlowerDifferential PrivacyPyTorchgRPC

Tech Arsenal

Machine Learning & AI

TensorFlow / Keras90%
PyTorch / PyG85%
Scikit-learn92%
MLOps & Pipelines80%

Programming Languages

Python95%
SQL88%
MATLAB85%
VBA / R75%

Data & Visualization

Pandas / NumPy93%
Power BI / Tableau85%
Advanced Excel90%
MySQL / SQL Server85%

Mathematics & Theory

Optimization90%
Probability & Statistics92%
Linear Algebra95%
Numerical Methods88%

Credentials & Certifications

Let's Connect

Get In Touch

I'm always open to discussing new opportunities, AI projects, or collaborations. Feel free to reach out!

Powered by Artificial Intelligence

Building the future, one model at a time