Background
I am a PhD student in Computer Science at EPFL, admitted to the EDIC Doctoral Program with a Fellowship. Previously, I graduated as valedictorian (ranked 1st) from Bilkent University's Computer Science program with a 3.96/4.00 GPA.
My current research focuses on uncertainty quantification and reducing LLM hallucinations to enable reliable use of large language models in high-stakes domains such as healthcare and scientific discovery. I am broadly interested in building robust, trustworthy AI systems that can reason through complex challenges where errors carry real consequences. Beyond CS, I draw inspiration from philosophy and economics — disciplines that shape how I think about decision-making under uncertainty, the ethics of autonomous systems, and the societal impact of AI.
Education
Ph.D. in Computer Science — EPFL
School of Computer and Communication Sciences (IC) · Sep 2025 — Present
EDIC Doctoral Program · EPFL Fellowship
B.S. in Computer Science — Bilkent University
Ankara, Turkey · Sep 2021 — Jun 2025
GPA: 3.96/4.00 · Valedictorian (1st/class)
Current Focus
My research centers on making large language models more reliable for high-stakes applications — through uncertainty quantification and hallucination reduction — contributing toward trustworthy AI for healthcare and scientific domains.
LiGHT Lab
Prof. Mary-Anne Hartley
Uncertainty quantification and reducing LLM hallucinations for high-stakes domains. Building methods to make LLM outputs more reliable and calibrated for clinical and scientific decision-making.
Rotation · Sep 2025 — Jan 2026IMOS Lab
Prof. Olga Fink
Developed an agentic AI framework for root cause analysis in cyber-physical systems through hypothesis-driven diagnostic reasoning with structured evidence accumulation. Also contributed to computational methods for mechanical bit sensors.
Research Interests
Uncertainty Quantification
Calibrating and quantifying uncertainty in LLM outputs for reliable decision-making in safety-critical applications.
LLM Hallucination Reduction
Developing methods to detect, measure, and mitigate hallucinations in large language models for scientific and medical contexts.
AI for Healthcare
Trustworthy AI systems for clinical decision-making, diagnostics, and healthcare applications where errors carry real consequences.
Reinforcement Learning
RL techniques for LLM alignment, safety, biological sequence optimization, and multi-agent coordination.
Agentic AI Systems
Multi-agent architectures for root cause analysis, knowledge graph construction, and scientific discovery workflows.
LLM Reasoning & Alignment
Complex reasoning capabilities, safety, and robustness of large language models for high-stakes scientific applications.
Research & Professional
Research Experience
Doctoral Researcher
LiGHT Lab — EPFL · Prof. Mary-Anne Hartley
Researching uncertainty quantification and methods to reduce LLM hallucinations for high-stakes domains. Developing calibration techniques and reliability metrics for LLM outputs in clinical and scientific decision-making.
Doctoral Researcher (Rotation)
IMOS Lab — EPFL · Prof. Olga Fink
Built an agentic root cause analysis framework that structures LLM reasoning through explicit hypothesis-driven diagnostic reasoning with log-odds belief tracking. Maintained belief scores over a discrete fault space, accumulating evidence across diagnostic steps for cyber-physical systems. Also contributed to a computational framework for mechanical bit readout through inverse dynamic programming.
Undergraduate Researcher
CICEKLAB — Bilkent University · Prof. A. Ercüment Çiçek
AI-driven drug discovery through transformer-based biological sequence generation. Pretrained and fine-tuned GPT-2 and T5 models for RNA sequence generation conditioned on target protein structures. Integrated GenerRNA and ProtBERT for protein-conditioned RNA generation. Applied Group Relative Policy Optimization (GRPO) with custom binding-affinity reward functions for RNA optimization. Validated outputs using DeepBind, GC content, and minimum free energy metrics.
Undergraduate Researcher
Semantic Communication Group — Bilkent University · Prof. Orhan Arıkan
Developed multi-agent reinforcement learning frameworks for wireless bandwidth allocation in city-vehicle IoT networks (collaboration with Huawei Ottawa). Designed goal-oriented semantic communication systems for cooperative driving using CARLA simulations. Investigated information theory approaches for efficient semantic signal processing. Presented findings to Huawei Wireless CTO Dr. Wen Tong.
Industrial Experience
AI Engineer
ROKETSAN
Built multi-agent LLM systems for tactical war game situational awareness. Designed agentic RAG pipelines with GraphRAG for knowledge extraction from domain documents. Developed multi-agent workflows for in-depth research report generation.
NLP Intern
Digital Transformation Office — Presidency of Turkey
Fine-tuned RoBERTa on 200K e-commerce reviews achieving 92.3% accuracy on rating prediction. Built T5-based summarization pipeline and cybersecurity data leak detection with Llama3 and DeepSeekCoder v2.
Candidate Engineer
ASELSAN — MGEO Elektrooptic
Developed console applications for camera systems and computer vision. Worked on secure communication through obfuscation techniques in the electro-optic systems division.
AI Research Intern
Huawei
Built CARLA traffic simulation environment with semantic cameras for multi-object scenarios. Trained Deep Q-Network (DQN) for autonomous vehicle control and decision-making.
Selected & Ongoing Work
Ongoing Projects
From Beams to Bits
Computational framework for mechanical bit readout through inverse dynamic programming. Formulates the readout problem as step-wise dynamic programming, recovering acceleration bounds for elastic beam mbits on rotating plates.
Modular Networks for Structured Reasoning
Developing modular network architectures that decompose complex reasoning tasks into structured, composable modules — enabling more interpretable and reliable LLM reasoning for scientific applications.
Selected Projects
Variant-Net Search Engine
Agentic search engine for genomics processing 300M+ nodes and 1B+ edges. Three search modes — Reasoning, Agentic (70+ tools), and Deep Research — powered by Graph RAG, Semantic RAG, and DGI pipelines.
Rubber Duck AI
Multi-agent learning platform combining Feynman Technique with Rubber Duck debugging. Three specialized agents — questioning, teaching, and assistant — interact over knowledge graphs built from user documents.
Multi-Agent Drone Simulation
LLM-based tactical system for autonomous drone missions. Hierarchical control with centralized GPT-4 strategist and decentralized agents for adaptive pathfinding and threat avoidance.
Bilkent Online
All-in-one student platform with marketplace, real-time messaging, community groups, and Bilkent GPT chatbot. Backed by Microsoft Founders Hub and OpenAI, live at bilkent.online.
Contributing To
Optinoe
AI-powered platform for manufacturing optimization. Agentic AI that thinks, acts, and predicts to eliminate downtime — featuring automated work orders, deep research engine, hybrid predictive maintenance, and digital twin simulation.
Project Reports
Agentic Root Cause Analysis via Hypothesis-Driven Diagnostic Reasoning
Enhancing RNA Generation for Protein Binding via Group Relative Policy Optimization
A Transformer-Based Framework for RNA-Protein Interactions: Integrating GenerRNA and ProtBERT
Cross-Modality Sequence Generation for Compound-RNA and Protein-RNA with Multilingual T5
Effective Communications in the Goal-Oriented Semantic Signal Processing Framework: A Case Study in Cooperative Driving
Honors & Scholarships
EPFL Fellowship
EDIC Doctoral Program admission with full fellowship
CS Valedictorian
Ranked 1st in graduating class — Bilkent University
Best Data Science Project
Variant-Net Search Engine — Bilkent CS492
1st Place — Amazon University Engagement
Rubber Duck AI · $4,000 scholarship
AWS AI-ML Scholarship
Amazon & Udacity scholarship program
ALES: Rank 28 / 139,565
Academic Personnel & Postgraduate Entrance Exam
Bilkent Full Scholarship
Comprehensive scholarship for academic excellence (2021–2025)
YKS'21: Rank 89 / 2,592,390
Higher Education Institutions Math-Science Exam
Community & Service
Bilkent AI Society — Co-Founder & General Secretary
Co-founded and led a 500+ member student society. Secured partnerships with NVIDIA and Notion. Hosted speaker events featuring researchers from OpenAI, Google, and Meta.
180 Degrees Consulting — Bilkent
Pro-bono consulting project for AÇEV (Mother Child Education Foundation) in collaboration with Bain & Company methodology.
Get in Touch
Open to collaboration and conversation.
Whether it's about research, a potential collaboration, or just an interesting idea you'd like to discuss — I'd love to hear from you. Feel free to reach out through any of the channels below.