🤓 Latest Submissions
.png&w=828&q=75)
Aegis
Aegis — The AI Firewall is an enterprise-grade security proxy and observability platform for AI agents. As organizations rapidly deploy AI agents in production, they face a critical blind spot: zero visibility into what those agents send to or receive from LLM APIs. One prompt injection, one PII leak, one data exfiltration attempt — and you have a compliance violation or breach. Aegis solves this by acting as a transparent reverse proxy between your AI agents and any LLM backend. Integration requires changing a single line of code — swap your LLM base URL to the Aegis proxy URL. No SDK, no agent modifications, no code changes. At its core, Aegis uses Veea's Lobster Trap binary for deep prompt inspection (DPI). Every request passes through 13 ingress firewall rules and 2 egress rules that detect prompt injections, PII/credential leaks, SQL injection, shell commands, data exfiltration, role impersonation, malware requests, and obfuscation attempts — all in sub-millisecond time using compiled regex patterns with zero LLM overhead. Google Gemini 2.0 Flash powers the intelligent AI responses that flow through the proxy, providing fast, accurate, and cost-effective completions for enterprise use cases. Key features include a real-time SSE-powered dashboard showing all agent activity, an agent registry with unique proxy URLs and configurable policy levels (strict/moderate/permissive), a built-in adversarial security tester with 27 attack prompts across 4 categories, full audit trails ready for SOC2/HIPAA/GDPR compliance, and multi-tenant user isolation. The platform is built with Python FastAPI, Next.js 16, SQLite, and deployed on Hugging Face Spaces (backend with Lobster Trap) and Vercel (frontend). It targets CTOs, Heads of AI, and Security Engineers at mid-to-large enterprises with a SaaS pricing model starting at $99/month.
19 May 2026
KYC Agent
KYC Agent is an intelligent, end-to-end Know-Your-Customer compliance platform built for Deriv. It replaces slow, manual document verification with AI-driven automation while keeping humans in the loop for high-risk cases. How it works: Users upload country-specific identity documents (CNIC, Emirates ID, Passport, Aadhaar/PAN) and proof-of-address through a guided multi-step onboarding wizard. Google Gemini 2.5 Flash performs real-time OCR extraction and document quality analysis — detecting 20+ issues like blur, glare, cropped corners, and potential manipulation. Extracted data is cross-validated against user-submitted form fields to catch mismatches and inconsistencies. Hybrid Risk Scoring Engine: A two-tier risk assessment combines deterministic rule-based checks (data mismatches, expired documents, missing fields, name inconsistencies with transliteration awareness) with Gemini-powered AI analysis for nuanced fraud detection. Submissions are auto-routed: LOW risk → auto-approved, MEDIUM → manual review queue, HIGH → flagged for expert review. Compliance Dashboard: An internal admin tool gives compliance officers a real-time queue of submissions with risk-sorted views, side-by-side form-vs-OCR comparison, detailed risk factor explanations, and one-click approve/reject actions with audit trail notes. Key differentiators: Country-aware intelligence with side-specific OCR (e.g., CNIC back = address extraction) Explainable risk scores — every flag comes with reasoning and severity Graceful Deriv WebSocket API integration with demo-mode fallback Production-grade architecture (FastAPI + Streamlit) with zero external infrastructure dependencies Built with: Streamlit, FastAPI, Google Gemini 2.5 Flash, Deriv WebSocket API, Pydantic, Python
7 Feb 2026
