Mohamed Adib@Adib05

Eufisky is a browser-based voice safety system designed to protect older adults from suspicious phone conversations without unnecessarily monitoring trusted contacts. Trusted callers bypass transcription and AI processing, while unknown callers are handled by a conversational Front Door voice agent that asks who they are and why they are calling. During an active call, Eufisky uses AssemblyAI Universal Streaming to generate low-latency transcripts for the connected audio legs. A deterministic, speaker-aware scam-risk engine evaluates signals such as urgency, threats, sensitive information requests, and known scam vocabulary. Risk thresholds trigger explainable escalation states, including alerts, a Guardian intervention, and family notification. After a call, Eufisky performs privacy-aware post-call processing, including PII redaction and incident summarization. The resulting information is stored in SQLite and presented through a Family Dashboard with live transcripts, evidence, risk scores, call history, and redacted incident reports. The architecture deliberately separates AI-generated conversation from deterministic safety policy, so the language model does not have unchecked authority over security-critical decisions. Eufisky is currently implemented as a browser phone simulation and is designed as a foundation for future real-telephony integration.

🌎 Online Hackathon | 💻 Month-long challenge | Dates: Sep 1–30, 2026 | 🏆 $10,000 Prize Pool ($5k cash + $5k in AAI credits) A month-long online challenge run by lablab.ai together with AssemblyAI, the Voice AI infrastructure company for builders. Every participant builds on AssemblyAI. You can join at any point — registration stays open for the whole build window, so you can start on day one or pick the project up halfway through and still submit. 🌍 Fully online hackathon. Join and build from anywhere in the world.