AI & Mobile Engineer · Istanbul, Turkey
Merve
Yuvacı
Computer engineer building intelligent systems at the intersection of AI and mobile. Award-winning developer, published app maker, and passionate about technology that genuinely changes lives.
my app StampUp — AI daily motivation, live on iOS
Recognition
Award-Winning
Work
Engineered a high-performance in-flight VR dashboard for Meta Quest. Built with Kotlin and Jetpack Compose, integrating real-time telemetry APIs with a custom Canvas-based 3D renderer for immersive spatial data visualization.
Architected a full-stack disaster response mobile app using Flutter and serverless Firebase. Integrated Computer Vision pipelines via Google Cloud Functions to auto-verify incident reports and synthesize real-time situational maps from user and drone data.
Career
Experience
- Reduced manual business calculation overhead and improved output accuracy by integrating OpenAI LLMs into core product workflows, applying context engineering strategies and domain-specific prompt design to automate complex computation tasks at scale.
- Minimized inference costs, hallucination rates, and response latency of production AI agents by designing evaluation pipelines to benchmark model behavior and drive iterative optimization.
- Delivered real-time data visibility across core product features by engineering FastAPI RESTful endpoints containerized with Docker, wired into a production full-stack architecture.
- Improved decision-making speed for cross-functional teams by shipping interactive React components that surfaced key business metrics directly within internal dashboards.
- Boosted data retrieval performance for high-traffic workflows by optimizing PostgreSQL queries and managing schema migrations.
- Accelerated development velocity by integrating Claude Code into daily engineering workflows, applying AI-assisted development practices across backend and frontend feature delivery.
- Strengthened codebase reliability by leading peer code reviews, writing unit tests, and driving performance optimization sprints across an active production codebase.
- Shipped production-grade conversational AI services in Python using LangChain, taking experimental prototypes to reliable, customer-facing products deployed in live environments.
- Architected graph-based decision automation workflows with n8n and NetworkX, eliminating manual intervention in complex multi-step data processing pipelines.
- Increased LLM answer accuracy and reduced hallucinations in production by architecting RAG pipelines with embedding models and MongoDB vector search, enabling semantic retrieval over verified enterprise knowledge bases.
- Cut LLM inference costs, reduced hallucination rates, and improved response latency across production AI agents by designing evaluation pipelines to benchmark model behavior and drive iterative optimization.
- Improved system reliability by driving debugging sessions and pull request reviews within Agile sprints, accelerating issue resolution across a remote-first engineering team.
- Safeguarded platform integrity for millions of users by analyzing 2M+ user-generated records to detect security anomalies and high-risk behavioral patterns using structured data analysis techniques.
- Translated raw content and behavioral data into actionable stakeholder reports through root-cause and trend analysis, directly informing product safety decisions at the leadership level.
- Accelerated bug resolution cycles within the SDLC by systematically triaging software defects surfaced through user feedback, reducing time-to-fix across multiple engineering teams.
Built
Projects
A published mobile app on the Apple App Store that uses Generative AI to create personalized growth roadmaps and custom digital "stamps" for user milestones. Full-stack Flutter development from UI/UX design to production-ready AI feature integration.
A physical autonomous robot using OpenCV's MobileNet-SSD for real-time object recognition. Dual-processor Raspberry Pi + Arduino architecture with camera and sensor arrays.
Award-winning disaster response platform. Computer Vision pipeline auto-verifies incidents; AI-driven map fuses user reports with drone data for real-time situational awareness.
High-performance in-flight VR dashboard for Meta Quest. Custom Canvas-based 3D renderer with real-time telemetry API integration and immersive spatial data visualization.
Attendance
Robust attendance tracking system with automated identity verification through facial recognition. Optimized for real-time, low-latency processing in production environments.
Intelligent travel planning powered by Google Gemini and Google Search, generating hyper-personalized, dynamic itineraries enriched with real-time map routes. Combines live search grounding with generative AI to produce contextually relevant, day-by-day travel plans tailored to user preferences.
Predictor
Designed and trained a machine learning model to forecast stock price movements, achieving ~60% prediction accuracy. Applied ML for financial time-series data analysis.
Toolkit
Skills & Technologies
Who I am
About
"Engineering intelligent systems that live at the edge of what AI and mobile can do."
I'm a computer engineering student at Beykent University (graduating 2026), currently working at Ivynest as a Software / AI Engineer. My focus is the intersection of AI and mobile — building things that are both technically rigorous and genuinely useful.
I won 1st place at the Meta Horizon Start Developer Competition ($25,000) for a VR AI flight dashboard, and placed 3rd at Samsung Innovation Campus with ECHO, an AI-powered disaster response platform. My app StampUp is live on the App Store.
Previously I built production GenAI services and RAG pipelines at Istvanslab, and spent 3 years at Brainly analyzing large-scale data for platform integrity.
I completed the Aspire Leadership Academy (Harvard-founded accelerator) and hold a specialization in AI & Deep Learning from Samsung Innovation Campus.
I volunteer with Make-A-Wish Turkey to fulfill the wishes of children with critical illnesses.
Let's connect
Open to exciting roles, collaborations,
and conversations about AI & mobile.