Loading repository data…
viochris / repository
🚀 High-performance RAG Backend for Silvio's Portfolio. Built with FastAPI, LangChain, FAISS, and Google Gemini to provide an interactive AI chat experience.
Silvio's Portfolio API is a next-generation interactive resume built on Retrieval-Augmented Generation (RAG) architecture.
Instead of reading a static PDF, recruiters and developers can chat with this API to ask specific questions about Silvio's experience, skills, and projects. It combines the reasoning power of Google Gemini with a precise Vector Knowledge Base (FAISS) to ensure every answer is factually grounded in Silvio's actual profile data, complete with source citations and relevance scores.
Unlike standard chatbots that hallucinate, this engine:
Trust but verify. Every response includes a source_list object that proves the AI isn't making things up. It provides:
0.51) indicating how closely the source matches the query.Built with LangChain's Conversational Retrieval Chain, the API remembers previous turns in the chat (via history parameter), allowing for follow-up questions like "Tell me more about that project."
silvio_profile.json.https://silvio0-silvio-portfolio-api.hf.space
Interact with the AI to learn about Silvio.
/assistantPOST{
"text": "Who is Silvio and what is his focus?",
"language": "English",
"history": []
}
{
"answer": "Silvio Christian Joe, known as Vio, is a highly driven 5th-semester Informatics Engineering undergraduate student at Universitas Dian Nuswantoro (UDINUS). He is an aspiring Data Scientist and AI Engineer based in Semarang, Central Java, Indonesia.\n\nVio specializes in **Natural Language Processing (NLP)** and **Tabular Data Analysis**, demonstrating a strong ability to transform raw data into meaningful insights. His expertise extends beyond just algorithms; he focuses on rigorous data preprocessing, effectively handling class imbalances using techniques like SMOTE/SMOTENC, and building robust, end-to-end solutions from model training to API deployment.\n\nHe is proficient in languages such as Python and SQL, and his data science toolkit includes libraries like Scikit-learn, Pandas, NumPy, Matplotlib, Seaborn, and NLTK. Vio is actively seeking internship opportunities to further apply and expand his skills in the field.",
"source_list": [
{
"doc_id": "28cf610e-3b82-4825-b625-1c6ddac96aa8",
"content": "{\n \"profile\": {\n \"name\": \"Silvio Christian Joe\",\n \"short_name\": \"Vio\",\n \"role\": \"Data Scientist & AI Engineer\",\n \"specialization\": \"Natural Language Processing (NLP) & Tabular Data Analysis\",\n \"university\": \"Universitas Dian Nuswantoro (UDINUS)\",\n \"current_semester\": 5,\n \"location\": \"Semarang, Central Java, Indonesia\",\n \"summary\": \"I am a 5th-semester Informatics Engineering undergraduate driven by curiosity to turn raw data into meaningful stories. I specialize in mining insights from structured numbers (Tabular) and unstructured language (NLP). My approach goes beyond algorithms; I focus on rigorous preprocessing, handling class imbalances (SMOTE/SMOTENC), and building end-to-end solutions from training to API deployment.\",\n \"availability\": \"Seeking Internship Opportunities\",\n \"quote\": \"Data is the new oil, but it’s useless without the engine to refine it.\"\n },\n \"contact\": {\n \"email\": \"viochristian12@gmail.com\",",
"source": "silvio_profile.json",
"score": 0.517580509185791
},
{
"doc_id": "3ff5bb20-ae54-443b-8bc4-4ab83496fbeb",
"content": "},\n \"contact\": {\n \"email\": \"viochristian12@gmail.com\",\n \"phone\": \"+62 895 3426 37871\",\n \"linkedin\": \"https://www.linkedin.com/in/silvio-christian-joe/\",\n \"github\": \"https://github.com/viochris\",\n \"address\": \"Jalan Bintoro VIIA No. 23A, Semarang\",\n \"cv_download\": \"https://github.com/viochris/viochris/raw/main/CV_Silvio_Christian_Joe_Data_Scientist.pdf\"\n },\n \"education\": [\n {\n \"institution\": \"Universitas Dian Nuswantoro (UDINUS)\",\n \"degree\": \"Bachelor of Informatics Engineering\",\n \"focus\": \"Data Science (NLP & Tabular Data)\",\n \"start_date\": \"August 2023\",\n \"end_date\": \"Present (Expected 2027)\",\n \"status\": \"Undergraduate Student\"\n },\n {\n \"institution\": \"SMA Kristen YSKI\",\n \"degree\": \"High School Diploma, Science\",\n \"start_date\": \"July 2020\",\n \"end_date\": \"May 2023\"\n }\n ],\n \"skills\": {\n \"languages\": [\"Python\", \"SQL\"],\n \"data_science\": [",
"source": "silvio_profile.json",
"score": 0.5361928939819336
},
{
"doc_id": "cfd2f84a-3848-4db2-970c-e98100384d03",
"content": "},\n {\n \"name\": \"Stuntify API\",\n \"domain\": \"API / MLOps\",\n \"role\": \"ML Engineer\",\n \"link\": \"https://github.com/viochris/Stuntify-API\",\n \"tech_stack\": [\"FastAPI\", \"Pydantic\", \"Scikit-Learn\"],\n \"description\": \"A high-performance inference engine serving real-time stunting predictions. Designed with strict Type Coercion, multi-artifact orchestration, and interactive Swagger documentation.\"\n },\n {\n \"name\": \"Insightify API\",\n \"domain\": \"NLP / API\",\n \"role\": \"Developer\",\n \"link\": \"https://github.com/viochris/Insightify-Sentiment-API\",\n \"tech_stack\": [\"FastAPI\", \"Transformers (RoBERTa)\", \"Pandas\"],\n \"description\": \"A dual-lingual (English & Indonesian) NLP microservice for sentiment analysis. Supports batch processing (Excel/CSV) and N-Gram keyword extraction.\"\n },\n {\n \"name\": \"DocuChat AI (Long Context)\",\n \"domain\": \"GenAI / LLM\",\n \"role\": \"Developer\",",
"source": "silvio_profile.json",
"score": 0.5770090818405151
},
{
"doc_id": "9a37ce03-7a8e-4804-9fec-49b68ec487f7",
"content": "\"issuer\": \"Coursera (IBM)\",\n \"year\": \"2025\"\n },\n {\n \"name\": \"Code Generation and Optimization Using IBM Granite\",\n \"issuer\": \"IBM\",\n \"year\": \"2025\"\n },\n {\n \"name\": \"Artificial Intelligence Essentials V2\",\n \"issuer\": \"Certiport / Relevant Body\",\n \"year\": \"2025\"\n }\n ],\n \"languages\": [\n {\n \"language\": \"Indonesian\",\n \"proficiency\": \"Native / Bilingual\"\n },\n {\n \"language\": \"English\",\n \"proficiency\": \"Professional Working Proficiency\"\n }\n ]\n}",
"source": "silvio_profile.json",
"score": 0.6120243668556213
}
]
}
Test the API directly in your browser without writing code:
POST /assistant.git clone https://github.com/viochris/silvio-portfolio-api.git
cd silvio-portfolio-api
.env file and add your Google Gemini API Key:GOOGLE_API_KEY=your_actual_api_key_here
pip install -r requirements.txt
uvicorn api:app --reload
Output: Uvicorn running on http://127.0.0.1:8000
Author: Silvio Christian, Joe "Data is the new oil, but it’s useless without the engine to refine it."
Selected from shared topics, language and repository description—not editorial ratings.
miracle5284 /
A showcase of my expertise in backend development, AI integration, and web technologies. This repo features scalable APIs, microservices, and cloud-native solutions built with Node.js, Python, Django, FastAPI, and React. Includes AI-driven automation, RAG-based retrieval, and optimized DevOps workflows for high-performance applications. 🚀