
5
3
Pakistan
1 year of experience
I am a student at Virtual University (VU) with a foundation in Python programming. I have knowledge and skills in machine learning (ML) and artificial intelligence (AI), and I am passionate about applying these technologies to solve real-world problems and grow my expertise in the field.

AURA is a voice-first safety co-pilot built for workers on active job sites, where hands are busy, gloves are on, and stopping to type a report isn't realistic. Instead of a form, workers just talk. AURA listens in real time, understands what's being reported, walks them through safety procedures step by step, and can flag hazardous conditions on the spot, like an out-of-range pressure or temperature reading, before they escalate. Under the hood, AURA uses AssemblyAI's real-time Voice Agent API for speech-to-text, conversational reasoning, and spoken replies. Reported hazards and near-misses go through a deliberate confirmation step before saving, so nothing gets submitted by accident, and every action is logged for audit. Once submitted, reports flow into a supervisor dashboard where safety officers can review incoming reports, approve or reject them, spot recurring patterns across sites, and track maintenance follow-ups. The result is a system where the person closest to the hazard can report it in seconds, without breaking focus on the task at hand, and the people responsible for site safety get a clear, real-time picture of what's happening in the field.
30 Sep 2026
.png&w=828&q=75)
Robots and self-driving machines need absolute certainty about their surroundings before they make a move. Our project acts as a real-time safety checkpoint for these autonomous systems. Instead of relying on a single point of failure, it constantly compares live data from multiple different sensors, including cameras, depth monitors, and lasers. Think of it as a team of spotters double-checking each other's work. If one sensor gets blinded by a glare, gets covered in dirt, or suddenly starts giving bad readings, our system catches the error immediately. It overrides the confused sensor, raises a high-risk warning, and completely stops the machine from making a dangerous physical movement. To show this in action, we built a live web dashboard. During our demonstration, viewers can use a slider to manually break a sensor reading on purpose. You can watch our system instantly spot the outlier, sound the alarm, and hit the brakes in real-time, proving that the machine only moves when it is objectively safe.
16 Sep 2026

FoundryAI is a multi-agent startup pitch generator that transforms ideas into complete, investor-ready presentations. The system integrates three specialized agents: PitchAgent, which generates structured startup pitches using the Mistral API; DesignAgent, which creates custom logos or slides via AIMLAPI; and VoiceAgent, which narrates the pitch using ElevenLabs. A Streamlit web interface enables users to input their ideas and instantly receive a pitch, visuals, and narration in one place. While the project references the Coral Protocol as a placeholder for future integration, it is not actively used in the current workflow due to SDK and time constraints. The codebase includes structure and commented sections for a CoralSession object, ensuring that the architecture is future-ready for Coral-based orchestration and agent communication. For now, all agents run independently as Python modules and communicate directly with external APIs. This approach allowed us to deliver a fully functional prototype within the hackathon timeline while leaving room for future Coral-based enhancements.
21 Sep 2025