Where developers grow together

ZEROTRACE

ZEROTRACE - AI Security Investigation Agent for SOC Analysts ZEROTRACE is a modular, evidence-grounded AI security investigation platform built for SOC analysts. It actively ingests security telemetry, correlates events into structured incidents, extracts IOCs, maps attacker behavior to the MITRE ATT&CK framework, and uses LLMs to synthesize factual, evidence-backed investigation reports and response actions. 🚀 Features Live Telemetry Ingestion: Normalizes raw logs (Sysmon, EDR, Network, DNS, Auth) into a unified SecurityEvent schema. Automated Incident Correlation: Groups related security events based on time windows and shared IOCs (hosts, users). MITRE ATT&CK Mapping: Automatically identifies attacker tactics and techniques from event types. Evidence-Grounded AI Synthesis: Generates human-readable incident summaries and remediation steps backed strictly by observed telemetry, without hallucinating details. SOC Analyst Console: An ultra-minimal, typography-driven React/Vite frontend tailored for high-speed triage. 🏗 Architecture Overview ZEROTRACE/├── .env                  # Environment configuration (API keys, DB URLs)├── backend/│   ├── app/│   │   ├── main.py       # FastAPI application entrypoint with CORS & API routing│   │   ├── core/         # Pydantic BaseSettings config & Database sessions│   │   ├── db/           # SQLite repository abstraction layer│   │   ├── models/       # Pydantic v2 schemas & SQLAlchemy ORM models│   │   ├── api/v1/       # REST API Endpoints (Incidents, Events, IOCs, Health)│   │   └── modules/      # Core Business Logic│   │       ├── ingestion/     # Log ingest and routing│   │       ├── normalization/ # Parsing and standardizing logs (DNS, Sysmon, etc.)│   │       ├── extraction/    # IOC extraction│   │       ├── correlation/   # Event clustering logic│   │       ├── mitre/         # Threat intelligence mapping│   │       └── investigation/ # AI and synthesis engines│   └── tests/            # Pytest test suite└── frontend/             # React (Vite) SOC Console    ├── src/    │   ├── components/   # UI Modules (Overview, Timeline, AI Investigation, etc.)    │   ├── services/     # API Axios client    │   └── index.css     # Minimalist Dark Theme    └── vite.config.js

Sentri AI - AI Voice Companion for Senior Citizens

SentriAI — Intelligent, Privacy-First Eldercare Companion & Safety Monitor SentriAI bridges warm daily companionship and reliable, instant emergency protection for elderly users. What it does Companion: A warm, full-duplex conversational voice assistant powered by Agora Conversation AI (via the Agora Voice/RTC SDK) that keeps the user company and checks in on them naturally. Reminder: On-device scheduling that guides users through daily medication, meals, and appointments. Emergency: Offline, on-device AI detects falls and verbal distress and instantly dispatches the emergency pipeline. Failsafe Mantra: If an elder in distress can’t explain their situation, repeating a simple predefined phrase (e.g., “Mimi Mimi Mimi” or “Mime Mime Mime”) 3+ times in a row bypasses the classifier and triggers the emergency pipeline. Emergency pipelineSentriAI lets the user configure an emergency helper contact (name and mobile number). When an emergency event is detected — a fall, verbal distress, or the failsafe mantra — the pipeline triggers automatically and sends a help message to that contact, including the user’s current location. Tech stack Frontend & Client: Native Android app in Kotlin + Jetpack Compose. Real-Time Communications: Agora Voice/RTC SDK for low-latency, full-duplex voice streams to the Agora Conversational AI Cloud Engine. Local Speech-to-Text: Whisper-tiny.en on PyTorch ExecuTorch for fast, local transcription. Local LLM Classification: Gemma 3 1B running locally via the Google MediaPipe LLM Inference SDK. Local Storage: Android Room Database for schedule caching and alert history. Backend Orchestration: FastAPI (Python) for serverless WebRTC token minters and Agora Studio agent configuration. Why it matters: Companion conversations run through Agora Conversation AI for rich, low-latency dialogue, while safety detection runs entirely on-device — so emergency protection is private, instant, and includes the user’s location the moment help is needed.

1D Simple Harmonic Oscillator (SHO) PINN

The project aims to develop a Physics-Informed Neural Network (PINN) to solve the 1D Simple Harmonic Oscillator (SHO) governing differential equation. The goal is to incorporate the underlying physics directly into the neural network’s loss function and accurately predict the oscillator’s displacement over time while satisfying the governing equation and initial conditions.