Aida Farahani, PhD | AI Solutions Engineer

About

I am an AI Solutions Engineer with a PhD in Artificial Intelligence, focused on building practical, reliable AI systems for real-world applications.

My work centers on industrial computer vision, data-centric AI, and end-to-end system design: from problem definition and dataset construction to model development, evaluation, integration, and deployment.

I specialize in improving model performance through data refinement and in designing hybrid systems that combine vision, geometry, and language models.

                        PhD in Artificial Intelligence
                        Technical University of Chemnitz
                    
                        Applied AI Focus
                        Industrial computer vision, data-centric AI, and production-oriented workflows
                    
                        Applied & Scientific Foundation
                        Practical AI for real-world systems, grounded in scientific machine learning and research methods

Core Expertise

Computer Vision

Object detection, instance segmentation, industrial image understanding, field-condition evaluation, and visual inspection workflows.

Data-Centric AI

Dataset design, annotation strategy, class balancing, failure-case analysis, and iterative model improvement.

AI Systems

End-to-end pipelines, workflow integration, production-oriented design, validation logic, and deployment planning.

Multimodal AI

Vision-language workflows: semantic retrieval, document/PDF understanding, and structured extraction from mixed media.

Deployment

CoreML, Swift/iOS integration, on-device inference, model export, and mobile visualization prototypes.

Scientific ML

FEM surrogate modeling, 3D deformation learning, mesh autoencoders, implicit representations, and RL for process modeling.

Skills

Python libraries (pandas, trimesh, plotly, meshlab, PyTorch Geometric) plus 3D and simulation workflows: mesh autoencoders, SDF, and FEM data.

Python libraries

PyTorch, TensorFlow, and scikit-learn for model development, plus computer vision: detection, segmentation, industrial imagery, preprocessing, and evaluation.

Frameworks

Docker, conda, Linux, Git, Jenkins, and Jupyter workflows; multimodal AI with RAG, CLIP, document/PDF understanding, and local LLMs (Ollama, Unsloth).

ML Tools

OOP and languages including C#.NET, C++, MATLAB, and Python; deployment with CoreML, Swift/iOS, on-device inference, and model export.

Programming Skills

Selected Projects

Multimodal Image Search

Retrieval · 2025

Built semantic image retrieval tools using language-driven search and similarity-based clustering.

Focus: CLIP-based embeddings, multimodal retrieval, image search, and clustering workflows.

CLIP Embeddings Search

Document & PDF Understanding

Vision-Language · 2025

Developed pipelines for extracting structured data from documents using vision-language methods.

Focus: document understanding, structured extraction, multimodal parsing, and workflow automation.

Documents PDF Multimodal AI

Data-Centric Model Improvement

Model Reliability · 2024 – present

Designed workflows to improve model performance through targeted data quality improvements.

Focus: annotation strategy, failure-case driven dataset updates, and iterative performance optimization.

Dataset Design Error Analysis Evaluation

Industrial Vision: Trench & Fiber Detection

Industrial AI · 2024

Designed segmentation pipelines for detecting trenches, ducts, and fittings in noisy field conditions.

Focus: instance segmentation for complex scenes, dataset refinement, class balancing, and field-condition evaluation.

Computer Vision Segmentation Data-Centric AI

On-Device Vision Inference

Deployment · 2024

Implemented real-time vision models on mobile devices with model optimization and iOS integration.

Focus: CoreML export, Swift/iOS integration, visualization, and field-ready inference prototypes.

CoreML Swift Mobile AI

Smart Airsense (BMBF)

Jan 2022 – 2024

Researcher on the BMBF project investigating AI methods for an interactive health assistant based on human-in-the-loop machine learning.

Focus: completed the first phase of a time series classification project using air particle sensor measurements to predict environmental events.

Time Series Sensor Data Predictive Modeling Human-in-the-Loop ML

Partners: Corant GmbH, TU Chemnitz.

3D Deformation Modeling (FEM Surrogate)

PhD Research · 2021 – 2023

Used reinforcement learning approaches for reverse engineering of metal forming process chains (prediction of actions for a desired deformed metal part).

Focus: sequential deformation modeling and reinforcement learning for process optimization.

Scientific ML FEM 3D Learning

ML@Karoprod (BMBF)

Feb 2018 – Jun 2021

Researcher on the BMBF project applying machine learning for process and quality prediction in automotive body production.

Focus: FEM-based dataset generation, deformation-property prediction, reinforcement learning for process modeling, and simulation-to-ML workflows for forming processes.

FEM Scientific ML Reinforcement Learning Automotive Production

Partners: Fraunhofer IWU Dresden, Scale GmbH, TU Chemnitz. Project code