I'm a Signal Processing and Imaging Engineer working across signal processing, computational imaging, and machine learning.

My core work sits in biomedical signal and imaging pipelines, from raw acquisition through reconstruction, quantitative analysis, and ML-driven classification and segmentation. The same toolkit extends into other domains: array processing and DOA estimation for acoustic sensing, control and state estimation for actuator systems, and digital communications processing. Whether the input is an OCT interferogram, a microphone array, or a sensor time series, the job is the same: turn raw measurements into reliable, interpretable systems, with classical DSP and machine learning used together where each fits best.

🔬 Core Expertise Biomedical signal processing and computational imaging Spectral-domain OCT and 3D volumetric reconstruction Numerical methods, inverse problems, and system modeling Scientific computing in Python, MATLAB, and C++ High-performance data processing and pipeline optimization Quantitative analysis and feature extraction from biomedical data

🔹 Spectral-Domain OCT 3D Pipeline
       Advanced reconstruction image and signal processing pipeline, including FFT, dispersion compensation, and volume processing

🔹 Digital PLL – I/Q Phase Detector
      Visual and mathematical exploration of digital PLL, NCO, and phase convergence analysis

🔹 Kalman Filter Actuator Simulation
      State estimation, system modeling, and simulation

🔹 Distributed Acoustic Source Localization
      Multi-array beamforming and DOA estimation (Bartlett, MVDR, MUSIC) with bearing fusion for 2D source localization

🔹 Vehicle Market Value Intelligence
      Data-driven pricing engine for used car inventory using regression analysis, OLS diagnostics, and feature engineering

🔹 Customer Behavior Prediction Engine
      Deep learning classification pipeline predicting customer return probability from 2 years of behavioural and engagement data

🔹 Biomedical Signal Clustering
      Pattern extraction, clustering, and feature analysis

🔹 Reinforcement Learning – Biped Robot Control
🔹 LLM-based Sentiment Analysis
🔹 CNN-based Image Processing Experiments
      Deep learning CNN framework in PyTorch for monocular depth estimation with metric-scale mapping.

• CUDA-accelerated signal processing kernels
• Memory-optimized FFT workflows
• Large-scale 3D biomedical volume handling

Languages

ML & Data

Imaging & Signal

Hardware & Interfaces

Dev Tools

• High-performance signal processing
• Research-to-production translation
• Applied ML for biomedical and sensing systems

• 💼 LinkedIn: https://www.linkedin.com/in/arman-rajaei/
• 📧 Email: arman.rajaei1992@gmail.com