
5
2
Indonesia
1 year of experience
Hello i'm Ridwan Yoga Suryantara. Undergraduate Information System and first year university student at Universitas Negeri Semarang. I have interest in Web development, Data Analyst, Android Developer (Kotlin), and Roblox Game Developer. Let's connect if you want!

Fuenzer Career is a comprehensive, voice-first AI interview coach designed to tackle a universal problem: 75% of job seekers feel unprepared for interviews, and traditional mock interviews are either prohibitively expensive or lack the realism of spoken conversation. Built entirely within the natively.builder ecosystem, Fuenzer Career provides a frictionless, guest-first experience—allowing users to start practicing for their dream role in under 10 seconds without creating an account. Our platform stands out by replacing static question banks with dynamic, market-driven intelligence. By integrating the Bright Data API, Fuenzer Career scrapes real-time job listings from platforms like LinkedIn and Indeed. This allows our AI agents to extract currently trending skills for any specific role and generate hyper-relevant interview questions that reflect what employers are actively seeking today. The core of the practice experience relies on a complex WebSocket integration with the Speechmatics API. Instead of typing answers or recording and uploading audio files, users speak naturally into their microphone. Speechmatics processes the raw audio stream in real-time, providing immediate transcription on the screen. This mimics the pressure and flow of a real interview. Once the interview concludes, our Evaluation Agent (powered by DeepSeek-V4-Flash via AI/ML API) analyzes the transcribed text against the scraped market data. It delivers a comprehensive dashboard featuring an overall score, a detailed skill-match breakdown, and actionable delivery tips. Crucially, the system analyzes the candidate's speech patterns, explicitly detecting "filler words" (like "um," "uh," or "like") to help users speak more confidently and professionally. Fuenzer Career demonstrates the power of combining real-time APIs, web scraping, speech-to-text, and large language models into a seamless, highly functional web application.
10 Aug 2026

The Problem: Today’s sports simulations are either overly complex video games that require hours of setup or boring statistical spreadsheets that lack engagement for casual fans. There is no platform that allows users to instantly test "what-if" scenarios using natural language or images. Our Solution: Fuenzer Sports redefines sports data interaction. Users can simply type a prompt (e.g., "Simulate the World Cup but give Argentina a +15 boost") or upload a picture of a tactical formation. Our Multimodal AI orchestrator interprets the request and triggers a highly optimized, vectorized Monte Carlo engine. In milliseconds, thousands of matches are simulated. The AI then generates an engaging narrative of the tournament in various personas, such as a TV commentator or a strict coach. Hardware & AI Architecture: Engineered for maximum throughput, our backend uses Python FastAPI and NumPy. We designed a dual-provider LLM architecture: a local pipeline configured for Google Gemma 4 on AMD Instinct MI300X (via ROCm 7.2) for future on-premise scaling, coupled with a dedicated Gemma 4 E4B production deployment on Fireworks AI for cloud resiliency. The Vision: Fuenzer Sports is a unicorn-trajectory platform aiming to empower sports broadcasters with live "what-if" TV analysis, provide betting platforms with generative predictive models, and give esports communities the ultimate custom tournament simulator.
13 Jul 2026