ML-powered_resume_analyser
Local, privacy-friendly resume analysis: convert, classify, and get advice using TF‑IDF, Logistic Regression, and sentence-transformer embeddings.
- 22
- Python
- 14 mo ago
tetratensor
☕ Making things as simple as possible...
Built from tetratensor's public GitHub profile, Japan. Not on Take My Repo yet, so there are no posts or updates here.
Local, privacy-friendly resume analysis: convert, classify, and get advice using TF‑IDF, Logistic Regression, and sentence-transformer embeddings.
A production-ready Copy Trading Platform Backend that enables automated trade replication from master to copier accounts across MetaTrader 4 and MetaTrader 5 platforms. Built with Node.js, Express, PostgreSQL, and Socket.IO for real-time operations.
Outbound PSTN calling agent using LiveKit SIP trunks with a voice pipeline (Silero VAD, Deepgram STT, OpenAI LLM+TTS). Includes a simple CLI for local development, health checks, and model prewarming.
Production-ready AI voice agent with LiveKit, OpenAI, ElevenLabs, Deepgram, and Twilio SIP integration. Features function calling, turn detection, noise cancellation, and inbound phone call support.
🎤 A real-time voice AI assistant built with Next.js and deployed on Cloudflare Workers. Features voice activity detection, speech-to-text, AI conversation, and text-to-speech with WebSocket communication and Durable Objects for seamless voice interactions.
Algorithmic trading client and strategies for a Jane Street–style electronic exchange: pennying market-making, micro-MACD, bond maker, ETF/ADR arbitrage.
Experimental web console for real-time multimodal AI interactions (voice/video) using Gemini 2.0 Live API and NijiVoice. Next.js (TypeScript, Tailwind) frontend and FastAPI backend communicate over WebSocket for low-latency streaming.
Devola is a Chrome extension for Amazon.co.jp that automatically displays expected reward points next to each item on your wishlist. No need to open product pages—see points at a glance with real-time updates and full client-side privacy.
"Development Rules" is a curated, practical guide to modern software engineering best practices. It covers coding standards, architecture principles, code review guidelines, testing strategies, documentation workflows, Git branching models, release management with feature flags, and team collaboration patterns.
The aim of this project is to create a real-time violence detection system using a custom-trained YOLO model. This system is designed to analyze video feeds, identify potential violent behavior, and issue alerts for immediate action.
Timelogger frontend built with Next.js + TypeScript. Track tasks and time with authentication, Storybook-driven UI, and typed OpenAPI client.
AI cat persona API with FastAPI, SSE streaming, and OpenAI integration
This project explores how U.S. economic metrics (such as GDP and Consumer Price Index) impact the performance of major world stock exchanges. Using Python and the FRED API, we retrieve economic data and compare it with stock price data obtained via the Yahoo Finance API.
A Python-based news sentiment analysis using Hugging Face Sentiment Analysis Model and Dataset.
A simple LSTM-powered sentiment analyzer for chatbots with a CLI and pretrained Keras model.
A simple JavaScript plugin that let users go back to the top of a web page.
Visual Polymarket wallet analyzer: activity patterns, bet sizing, entry/exit prices, PnL
Lightweight Python toolkit for multimodal speech/text workflows: STT, NER anonymization, sentiment, QA, doc→speech, and text generation using VOSK and Hugging Face.