A closed-loop security runtime preventing "The Great Exfiltration" and Indirect Prompt Injection in Autonomous AI Agents.
-
Updated
May 28, 2026 - Python
A closed-loop security runtime preventing "The Great Exfiltration" and Indirect Prompt Injection in Autonomous AI Agents.
A systems research project exploring how Zig changes the design of dynamic runtimes as a high-level Python implementation.
Linux network namespace-based transport performance benchmarking framework using tc, netem, iperf3 and optional eBPF instrumentation.
Experimental Linux RFC for an HBF/CXL-era AI memory control plane: runtime hints, prefetch, placement, and tiering.
Research harness for evaluating query-time bounded elimination of reconstructable KV-cache witnesses in long-context transformer inference workloads. Related provisional filing: IN 202641062451.
Experimental Linux kernel patchset and benchmark suite for semantic memory hints in inference workloads. Explores whether user-space intent (streaming vs reuse vs ephemeral memory) can influence reclaim behavior in Multi-Gen LRU (MGLRU).
Trace-driven research harness for KV-cache hierarchy policy evaluation in long-context LLM inference.
Exploratory AI infrastructure project modeling semantic KV-cache orchestration, memory tiering, and HBM/CXL movement tradeoffs for long-context LLM inference.
Runtime-core research for long-lived AI surfaces: worker ownership, transaction scheduling, and bounded projection.
A systems research platform for semantic KV-cache orchestration, topology-aware memory placement, distributed prefix reuse, and rack-scale inference memory simulation.
Simulation study of cache architecture tradeoffs under concurrency — partitioned vs LRU vs client affinity, with trace-driven evaluation on Twitter cache workloads
Undergraduate systems research paper comparing execution efficiency and system accessibility between native compiled and browser-based runtime environments using C, Python, and JavaScript.
Deterministic stability framework for stateful AI recovery under bounded compute with nonlinear collapse analysis.
Frictionless Computing: Entropy-Based Operation Filtering for 10x-1000x Effective Speedup
Runtime for survivable autonomous software agents using WASM, migration, and runtime economics.
Add a description, image, and links to the systems-research topic page so that developers can more easily learn about it.
To associate your repository with the systems-research topic, visit your repo's landing page and select "manage topics."