Ph.D. Candidate · Data Scientist & AI Researcher
Models get judged on data that looks like their training set. Deployed systems rarely get that luxury. I work at that failure point: robust under drift, explainable under scrutiny, small enough to run at the edge.
Istanbul, Türkiye
Medical imaging, wireless sensing, video quality: one problem from three angles. What happens to a model when reality stops matching the training set?
If I train on one device today, why does it fail on another device tomorrow?
Hardware differences and temporal drift quietly break the i.i.d. assumption almost every architecture rests on. I cancel that divergence at the signal level rather than papering over it with more parameters.
Would a clinician stake a decision on this prediction?
Accuracy is not evidence of understanding. I pair predictions with gradient-based attribution, so a specialist can see why the model decided and catch the cases where it is right for the wrong reason.
Does it still work on the hardware that actually exists?
A model that needs a datacenter is useless on a rescue robot. I report latency as a first-class result and use sparse Mixture-of-Experts routing to buy capacity without paying for it at inference.
Written the way I would present them to a research group: the question, the method, the numbers, and what did not work.
Wi-Fi fingerprinting collapses when you test it on a different day than you trained it. Rather than discard corrupted packets, this framework restores the physical signal and recovers saturated receivers.
CNNs read texture but miss the global context radiologists rely on. Pairing convolutional locality with transformer attention closed that gap.
Two receivers in the same spot report different RSSI. Deep feature learning over the TUJI1 dataset recovers a device-invariant representation.
Two Wi-Fi captures I recorded at the same location, 19.4 minutes apart. Nothing moved but the clock, yet a classifier can tell them apart. Computed live in your browser.
Reference: ref_204_1_3500.pcap (3500 packets) · Test: test_204_1_100.pcap (100 packets) · 52 usable subcarriers · PCA basis fitted on the reference session only.
How to read this: the session discriminator guesses which capture a packet came from. 50% means it cannot tell, which is the goal. High accuracy means a localizer trained here learns when the data was captured, not where the device is.
CORAL whitens the test session and re-colours it with the reference covariance, X′ = (X−μt)·Σt−1/2·Σs1/2 + μs, using no labels from the test session.
Running, maintained, and answerable to someone other than a reviewer.
A trade-show lead pipeline that turns a pile of business cards into clean, queryable records. Gemini Flash 2.5 reads each card directly, and the SQLAlchemy/PostgreSQL schema resolves duplicates on ingest rather than during a later cleanup.
A donation platform for orphan families in Sudan, built so donors never have to take the word "distributed" on trust. Every transfer is verified, converted at a recorded rate and disbursed against an official receipt, with per-family timelines published openly.
A corporate platform for a medium- and low-voltage switchgear manufacturer, built around one route: unfamiliar visitor to qualified quote request. Parallel Turkish/English routing, global search, and KVKK compliance handled up front.
Five papers. Each with a plain-language note on why it matters.
Telecommunication Systems (Springer), 2026
IEEE Access, 2025 · doi:10.1109/ACCESS.2025.3621505
@article{saeed2025mitigating,
author = {Saeed, M. T. M. and Yousif, M. A. A. and Ozturk, I.},
title = {Mitigating Device Heterogeneity for Enhanced Indoor Positioning
System Performance Using Deep Feature Learning},
journal = {IEEE Access},
year = {2025},
doi = {10.1109/ACCESS.2025.3621505}
}
Electronics (MDPI), 2026, 15(8), 1666 · doi:10.3390/electronics15081666
@article{dutta2026impact,
author = {Dutta, A. and Saeed, M. T. M. and Arawade, S. and Sam{\v{c}}ovi{\'c}, A.
and Uddin, S. and Juszka, D. and Grega, M. and Leszczuk, M.},
title = {Impact of Environmental Control on Subjective Video Quality
Assessment in Crowdsourced QoE Experiments},
journal = {Electronics},
volume = {15},
number = {8},
pages = {1666},
year = {2026},
doi = {10.3390/electronics15081666}
}
2024 12th Int. Scientific Conference on Computer Science (COMSCI), Sozopol, Bulgaria · doi:10.1109/COMSCI63166.2024.10778503
@inproceedings{saeed2024transconvnet,
author = {Saeed, M. T. M. and Yousif, M. A. A. and Hakim, A. A. M.},
title = {TransConvNet: Enhancing Kidney Abnormality Detection in CT Imaging
through Hybrid Transformer-CNN Model with Integrated Explainability},
booktitle = {2024 12th International Scientific Conference on Computer Science (COMSCI)},
address = {Sozopol, Bulgaria},
pages = {1--5},
year = {2024},
doi = {10.1109/COMSCI63166.2024.10778503}
}
Control engineering first, machine learning second. Model the physics, then learn the residual: that instinct comes from the former.
AGH University of Science and Technology · Kraków, Poland
TÜBİTAK & Osmaniye Korkut Ata University · Türkiye
AGH University of Science and Technology · Kraków, Poland
Al Neelain University · Khartoum, Sudan
Amsterdam, Netherlands
Split, Croatia · sponsored by Google DeepMind
Fully funded by Mastercard
Ondokuz Mayis University · Samsun
Istanbul University-Cerrahpaşa
Al Neelain University · Khartoum
I'm looking for roles and collaborations where the data is messy, the deployment is real, and the failure modes matter.