Projects
Active2025 – present
NIKA
Network Incident Benchmark for AI Agents
NIKA (Network Incident Benchmark for AI Agents) is an open platform for benchmarking AI agents on realistic network troubleshooting tasks—think SWE-Bench, but for network engineers.
It deploys live emulated networks (Kathará, Containerlab), injects realistic faults, exposes MCP tools and telemetry, and evaluates agent submissions under reproducible conditions.
Related news
Sep 1, 2026
NIKA public leaderboard and nika-bench 0.2.0We published nika-bench 0.2.0 and opened a public leaderboard for comparable agent runs on network troubleshooting.
Jul 7, 2026
NIKA project website is liveThe NIKA project site is public, with overview, architecture, and links into the benchmark and code.
Dec 18, 2025
NIKA preprint releasedWe released a preprint on NIKA, a Network Incident Benchmark for AI Agents on network troubleshooting.
Jul 1, 2025
Paper accepted at ACM SIGCOMM 2025 NGNO WorkshopOur lightning paper on a playground for AI agents in network troubleshooting was accepted at SIGCOMM’25 NGNO.
Active2024 – present
ChamaleoNet
Programmable passive probe for visibility on erroneous traffic
ChamaleoNet turns production networks into telescopes for erroneous and unsolicited traffic—capturing internal radiation that classic darknet monitors miss.
It combines P4 in-hardware filtering, a C-based VNF for buffering and anomaly handling, and DPDK anonymization, and has been running on the Politecnico di Torino campus network for over 1.5 years.
Related news
Jan 15, 2026
ChamaleoNet accepted at IEEE Transactions on NetworkingOur programmable passive probe for erroneous traffic visibility appears in IEEE ToN, with open code and an arXiv preprint.
Jan 10, 2026
EAGLE to appear at APNet 2026EAGLE applies graph representation learning to erroneous traffic collected by ChamaleoNet-style telescopes.
Oct 28, 2025
IMC 2025 poster on silent internal anomaliesAt ACM IMC’25 we presented a poster on uncovering silent internal anomalies from erroneous outbound traffic.