Hi, my name is
Eray Alp.
I’m a Software Engineer specializing in building intelligent, scalable, and production-ready systems. With a strong foundation in Machine Learning and NLP, my experience spans both Engineering and Applied AI. I focus on applying modern AI techniques to solve real-world problems.
01.About Me
I am a Software Engineer and Machine Learning Scientist with a passion for building robust, intelligent systems. My experience bridges applied machine learning, natural language processing, and modern software engineering.
I focus on both building scalable systems and doing AI research. I work on everything from production ML pipelines and backend services to training language models, researching search and recommendation systems. I enjoy combining solid engineering with machine learning research.
I’m currently an AI Research Engineer Intern at sahibinden.com, where I develop and implement AI solutions for one of Turkey’s largest online marketplace platforms. Previously, as a Machine Learning Engineer Intern at Insider, I worked on production-scale semantic search pipelines, fine-tuning transformer models, and optimizing retrieval systems.
Earlier in my journey, I gained complementary experience as a DevOps Engineer Intern at P.I. Works, contributing to internal AI agent development with LLMOps workflows and managing Kubernetes infrastructure. I also conducted research at COSMOS AI Research Group, Yıldız Technical University, where I fine-tuned Turkish LLMs for mathematical reasoning with reinforcement learning. My early experiences include working on data analysis at Istanbul Bilgi University and building full-stack e-commerce features at Insider as a Software Developer Intern.
Some of the tools & technologies I enjoy working with:
- Python
- PyTorch
- FastAPI
- Docker
- Kubernetes
- AWS
- Go
- TypeScript
- Vue
- Linux
- Git
02.Experiences
- Sep 2025 – Present
AI Research Engineer Intern@sahibinden.com ↗
Istanbul, Turkey
- Transitioned a rule-based moderation pipeline to an LLM-based system for Turkish user messages, improving semantic violation detection (advertising, fraud, toxicity, off-platform contact) and reducing false negatives by 35%.
- Developed a fast and scalable LLM inference service on GCP using Pub/Sub for asynchronous message processing, enabling low-latency moderation and reasoning generation over 150K–200K user messages daily.
- Fine-tuned smaller open-source LLMs using weakly supervised user message data to achieve comparable moderation performance to proprietary APIs while reducing inference cost.
- Automated dataset versioning and annotation checks via CI/CD and cloud triggers, ensuring data integrity across releases.
- Integrated evaluation and monitoring dashboards to track latency, drift, and violation trends for continuous improvement.
- Jul 2025 – Sep 2025
DevOps Engineer Intern@P.I. Works ↗
Istanbul, Turkey
- Contributed to internal DevOps AI agent development with LLMOps workflows, leveraging Retrieval-Augmented Generation (RAG), prompt engineering, and vector search to analyze commits and pipeline failures.
- Built and maintained CI/CD pipelines using GitLab CI/CD, Bash, and Python on Linux environments, automating dependency scanning, version tracking, and quality checks for greater reliability.
- Deployed and managed containerized applications on bare-metal Kubernetes clusters.
- Managed infrastructure observability through monitoring and alerting with Prometheus and Grafana, and centralized logging via the ELK Stack (Elasticsearch, Logstash, Kibana).
- Jan 2025 – Jun 2025
AI Research Assistant@COSMOS AI Research Group, Yildiz Technical University ↗
Istanbul, Turkey
- Selected for the TÜBİTAK 2247-C STAR Scholarship Program. Conducted research under the supervision of Prof. Mehmet Fatih Amasyalı.
- Fine-tuned Turkish Large Language Models (LLMs) to improve mathematical reasoning capabilities, curating and organizing custom datasets aligned with research objectives.
- Applied Reinforcement Learning (RL) to enhance reasoning and decision-making in NLP tasks, delivering a 20% improvement on Turkish grade-school math benchmarks with a compact model.
- Aug 2024 – Dec 2024
Machine Learning Engineer Intern@Insider ↗
Istanbul, Turkey
- Supported the search team in building a high-scale, multi-indexed search platform for improved search relevancy and personalized e-commerce results.
- Contributed to batch and streaming pipelines for data ingestion, daily embedding extraction and large-scale vector indexing using AWS Lambda, SQS, S3, Kinesis Firehose, and Redis.
- Assisted in testing hybrid retrieval methods and reranking models for better multilingual search relevance in production.
- Contributed to the Faceted and Banner Merchandising features with FastAPI, Elasticsearch, and Spring Boot, enabling customizable search experiences for partners.
- Gained hands-on experience in ML workflows and model deployments using Airflow and SageMaker, and optimizing microservices on AWS, Kubernetes, and Docker.
- Configured and provisioned AWS EC2 instances using Terraform and Terragrunt to support internal tool development.
- Mar 2024 – May 2024
Big Data Research Assistant@Istanbul Bilgi University, Center for Financial Studies ↗
Istanbul, Turkey
- Contributed to an economic research project at the Center for Financial Studies (CEFIS) by scraping and analyzing large-scale inflation data, applying time series modeling for trend prediction, and automating pipelines with Scrapy, Playwright, Docker, and Cron to ensure efficient, reliable data collection.
- Jul 2023 – Sep 2023
Software Developer Intern@Insider ↗
Istanbul, Turkey
- Built full-stack e-commerce personalization features with Node.js, developing REST APIs, scalable backend services for behavior tracking and customer segmentation, and contributing to CI/CD pipelines and codebase refactoring to ensure scalable, high-traffic platform performance.
03.Contact Me
Get In Touch
If you want to talk about opportunities and collaboration, my inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!