Selected academic curriculum vitae
Elgun Hasanov
MSc · Data and Artificial Intelligence
Institut Polytechnique de Paris · Paris, France
Research interests
My research focuses on machine-learning evaluation, large language models, formal methods, and clinical machine learning.
Education
2025 — 2027 (expected)
MSc, Data and Artificial Intelligence
Institut Polytechnique de Paris
École Polytechnique & Télécom Paris. Expected graduation: July 2027. French Government Excellence Scholarship (BGF).
2022 — 2025
BSc, Computer Science
Université de Strasbourg
Valedictorian of the graduating class.
Research experience
Jan — Apr 2026
Student Researcher · NLP & Graph Methods
Télécom Paris · Institut Polytechnique de Paris
Semantic alignment and retrieval across MathComp and mathlib, combining language embeddings, lexical retrieval, and graph-based refinement.
DIG team, LTCI · Supervised by Prof. Thomas Bonald.
May — Sep 2025
Research Intern · Computer Vision & Deep Learning
FEMTO-ST Institute (CNRS)
Self-supervised learning for holographic wavefront reconstruction, with physics-informed losses and comparative model evaluation.
AND team, DISC · Supervised by Prof. Raphaël Couturier.
Jan — May 2025
Research Intern · Unsupervised Learning
ICube · Université de Strasbourg
Analyzed 10,597 fish-school observations from marine acoustic data using clustering and ensemble methods. Compared K-Means, DBSCAN, HDBSCAN and Gaussian mixture models; HDBSCAN reached a silhouette score of 0.655.
Final-year bachelor’s research internship in the IMAGeS team.
Industry experience
Mar 2023 — Present · Baku · part-time
Co-Founder & Applied AI · Codage
RAG & Voice AI
Built FraudShield AI for financial fraud detection, using RAG over transaction history with FAISS, sentence-transformers, an LLM reasoning layer and few-shot chain-of-thought prompting. Built the MySA voice payment assistant with fine-tuned Whisper speech recognition and a text-to-speech pipeline.
Part-time alongside studies and internships; relocated to Paris for the IP Paris MSc.
Apr — Sep 2026
Applied AI Engineer · AI To Market
LLM Fine-Tuning & Evaluation
Jun 2024 — May 2025 · Baku
Data & AI Architect Intern · PASHA Life
Multi-agent AI & Data Platforms
Built a multi-agent AI system with speech-to-text, RF/XGBoost segmentation and Microsoft Fabric analytics. Designed an APISIX API gateway handling 50K+ daily policy requests at under 40 ms P95 latency, plus MDM and data-lake work with Azure Synapse, Snowflake, Kafka, Kubernetes and AWS Lambda.
Publications & preprints
Primary-Preserving Complementary Module Retrieval Across Formal Mathematics Libraries
Accepted for proceedings · Formal mathematics / semantic retrieval
Asymmetric Cross-Cohort Generalization in Alzheimer's Disease Classification: A Bidirectional External Validation Study
Accepted full paper · Biomedical AI / external validation
Leakage-Safe Machine Learning for Alzheimer's Disease Classification with External Validation on ADNI and OASIS
Accepted for oral presentation and proceedings · Paper 337
AttentionDep: Knowledge-Infused Attention for Interpretable Depression Severity Assessment from Social Media
Accepted for presentation · Mental-health NLP / interpretable attention
Can We Trust LLMs for Mental Health-Based Decisions? A Causality Aware Reliability Analysis
Published · Open access · LLM reliability / causal analysis
Decision Support through Feature-Aware Ensemble Clustering: A Multi-Algorithm Fusion Framework for Complex Data Analysis
Published · Ensemble clustering / decision support
Rapid Automated Alzheimer's Disease Classification Using FastSurfer Segmentation and Explainable Machine Learning
Published · Medical imaging / explainable ML
AttentionDep: Domain-Aware Attention for Explainable Depression Severity Assessment
Preprint · Earlier version of the WISE 2026 AttentionDep paper
Honors & awards
Stanford ASES France Fellowship
HEC Paris deep-tech finalist among 300+ teams
French Government Excellence Scholarship
Awarded for my MSc studies at Polytechnique Paris
Google AI Hackathon — 1st Place
FraudShield AI detects financial fraud with RAG
GreenTech ClimAccelerator 2024 — 1st Place
Ecoffin pairs engine fuel with a CO₂ dashboard
PASHA Hackathon 2.0 & 3.0
Invited as Innovation Mentor for 4.0, 5.0 & 6.0
TEKNOFEST T3 AI Hackathon
Fine-tuned LLMs for Turkish-language understanding
11+ hackathon wins
Across multiple competitions
Certifications
Oracle Cloud Infrastructure 2025 Certified Data Science Professional
1Z0-1110-25
GInI Certified Design Thinking Professional®
CDTP®
Oracle Cloud Infrastructure 2024 Data Certified Foundations Associate
1Z0-1195-24
Oracle Cloud Infrastructure 2024 Certified AI Foundations Associate
1Z0-1122-24