DSML Kazakhstan
Personal profile
Aimira Kazybayeva
@Amy
Data Scientist / Product Analyst · QazCode
I'm an ML Engineer at ForteBank in Astana, where I build production machine learning systems for a commercial bank.
About
My work spans fraud detection (I built a dropper-detection model with LightGBM/XGBoost that improved F1 from ~0.49 to ~0.66 by adding behavioral session features), customer segmentation (KMeans/DBSCAN clustering across millions of retail clients, deployed as a Dash service), and personalized deposit pricing with A/B testing. Before ML, I spent four years at SLB (Schlumberger) as a Field Engineer and later as Product Owner of an internal operational platform — so I've seen both the operational and technical sides of building something that actually gets used. I studied Mechanical Engineering (BEng) at the University of Bristol, and I'm trilingual in Kazakh, Russian, and English.
How I can help
- Hands-on experience taking ML models from prototype to production in a regulated industry (banking) — the messy parts most tutorials skip: data pipelines (Airflow), serving (FastAPI/Docker), and monitoring - A cross-domain perspective from engineering + product + ML, useful if you're trying to translate a model into something a business actually adopts - Practical know-how in fraud/anti-fraud modeling, customer segmentation, and behavioral feature engineering - Happy to review resumes/cases, share notes on ML interview prep (I've been through quite a few technical rounds recently), or just talk through a stuck problem
Looking to connect
- Connections with like-minded ML/data people in Kazakhstan — people building real things, not just talking theory - A mentor, ideally someone senior in ML engineering or a related field, who could offer guidance as I grow in my career - Exposure to how others are solving similar problems (fraud, segmentation, pricing) outside of banking, and general opportunities to learn from more experienced practitioners