Drishti AI — Network risk intelligence scanner
AI network scanner that visualizes attack paths and generates security remediation playbooks.

Live demo
The running product, embedded below. If it refuses to load in an embed, the live site opens in a new tab.
Problem
Network audits drown stakeholders in jargon: pentesters find paths an attacker would take, but can't explain the dollar blast radius — so fixes get deprioritized.
Solution
Drishti AI scans network ports and reachability, models the attack graph, prices the risk in dollars, and drafts remediation playbooks with an interactive Spline 3D mesh and Groq LLaMA 3 inference.
Decisions I made
- 01Price the risk in dollars — 'how safe is it' has no numeric answer and loses to questions that do; a priced attack path converts a CVV-level finding into a budget decision.
- 02LLM as translator, not scanner — the model drafts remediation over a deterministic attack graph and never invents findings; the scanner stays the source of truth.
- 03Graph, not port list — an open port on an internal print server isn't a risk; reachability between assets is what turns findings into attack paths a viewer can follow.
- 043D visualization for the non-technical room — the Spline mesh exists so a budget holder can see the path, not read about it; that's the whole point of the handoff.
Outcome
Board-ready security reporting: critical attack paths surfaced, priced, and paired with auto-compiled fix plans — the same core later won an innovation final at InnoFusion 3.0.
This project is part of the work I ship as a full-stack developer — see the freelance services page for project types, process, and how to start.