Works

Fintera

AI Product · Financial Intelligence · On-Premise

An on-premise financial intelligence platform that turns company financial documents into structured reports, consolidated views, and source-backed executive insights. Fintera is designed for teams that work with sensitive financial data and cannot send documents to third-party AI APIs or external cloud systems.

Problem: Finance teams often work across Excel files, PDFs, ERP exports, and disconnected reports. The challenge was to make this data searchable, explainable, and useful for decision-making without moving sensitive documents outside the company environment.

Contribution: Designed and built the product architecture for document ingestion, financial data extraction, consolidation workflows, anomaly detection, and natural-language reporting. Focused on a privacy-first on-premise setup, source-linked answers, and a workflow where AI supports finance teams without becoming an unverifiable black box.

Outcome: A productized financial intelligence platform with local deployment, document-aware reporting, explainable AI outputs, and a clear B2B positioning for finance-heavy companies.

Stack: Next.js · Node.js · PostgreSQL · Docker · RAG · Local LLM · Document Processing

Fintera ↗ (opens in a new tab)

Kapalıçarşı

Mobile App · Financial Education · Live on the App Store

A mobile-first financial education platform that helps users learn markets through virtual trading, portfolio tracking, and gamified competition. The product combines real-time market data with an approachable mobile experience, so users can explore assets like stocks, gold, crypto, and foreign exchange without risking real money.

Problem: Make financial learning more practical and engaging than static educational content, while keeping market data, portfolio state, and user progression understandable on mobile.

Contribution: Designed and developed the core mobile product, including virtual trading flows, portfolio tracking, live market screens, educational modules, gamification, leaderboards, and data-backed asset views. Built the supporting backend architecture for market data, user progress, and app state.

Outcome: A live App Store product with real-time financial data, simulated trading, learning mechanics, and competitive progression.

Stack: Swift · Go · PostgreSQL · Redis · Supabase · Next.js

Kapalıçarşı App Store ↗ (opens in a new tab)

TetryLabs

Brand Website · Lead Generation · Product Studio

A B2B product studio website built to position TetryLabs around custom software, AI automation, product development, and technical delivery. The goal was not just to publish a portfolio page, but to create a clear acquisition channel for founders and growing companies that need senior technical help.

Problem: The previous positioning was too generic for client acquisition. The site needed to explain what TetryLabs does, who it is for, why it is credible, and what the next step should be without feeling like a template agency website.

Contribution: Designed and built the website structure, service positioning, bilingual copy, selected work pages, lead magnet flow, contact form, SEO/GEO assets, analytics, and Google Ads conversion tracking preparation. Focused the messaging around business problems, technical clarity, real product delivery, and low-friction lead capture.

Outcome: A production website with clearer positioning, stronger trust signals, selected case studies, measurable lead events, and a conversion-focused path from visit to discovery call.

Stack: Next.js · TypeScript · Tailwind CSS · GA4 · Google Ads · SEO · GEO

TetryLabs ↗ (opens in a new tab)

Event Ticketing Platform

SaaS · High Traffic · Under NDA

A high-traffic ticketing platform built for event discovery, seat selection, reservation, payment flows, and operational management. The system needed to handle short bursts of demand, protect seat inventory during checkout, and give internal teams reliable tools to manage events, venues, ticket categories, and sales activity.

Problem: Prevent double-booking and checkout conflicts during high-demand ticket sales, while keeping the customer experience fast and the admin workflows manageable.

Contribution: Worked across the platform end to end: Go microservices for real-time seat reservation, Redis-backed locking and availability flows, a NestJS backend for authentication, CMS, payments, and operational APIs, plus Next.js applications for the customer storefront and admin dashboard.

Outcome: A scalable ticketing system foundation with real-time reservation logic, admin-controlled event operations, and customer-facing purchase flows designed for traffic spikes.

Stack: Go · NestJS · Next.js · PostgreSQL · Redis · Kafka

PatiAtlası: Kedileri Yakala

Mobile Game · Location-Based · Live on the App Store

A real-world cat discovery game where players find, capture, and collect cats around them. I built the native iOS app with location-based discovery, camera capture, AI-assisted verification, rarity tiers, XP, leaderboards, and shareable catch cards.

Problem: Keep community sightings trustworthy and prevent fake captures.

Contribution: Built the SwiftUI app and server-side AI verification flow with Supabase Edge Functions and AWS Rekognition.

Outcome: A live App Store product with real-world gameplay, collectible progression, and cheat-resistant verification.

Stack: SwiftUI · Supabase · AWS Rekognition · Apple Sign In

PatiAtlası: Kedileri Yakala App Store ↗ (opens in a new tab)

Private Social Video Platform

Mobile App · Video Chat · Under NDA

A native iOS video chat product built around user onboarding, real-time discovery, online matching, and low-latency video calls. The product required a smooth mobile experience, reliable call setup, and a backend flow that could handle user presence, matching states, and session lifecycle without making the app feel heavy.

Problem: Create a fast, trustable video matching experience where users can sign up, find available people, and enter calls with minimal friction.

Contribution: Built the native iOS experience and contributed to the backend architecture for authentication, online user matching, call session handling, and real-time communication flows. Integrated video infrastructure with Agora.io and AI-enabled call workflows using LiveKit.

Outcome: A production-oriented mobile product currently in review, with real-time matching, video call readiness, and a backend foundation designed for low-latency social interaction.

Stack: Swift · Go · Agora.io · LiveKit