Anoneurx Lab
Independent research into cybersecurity, operating systems, autonomous robotics, defense technologies, and advanced surveillance systems.
Robotics
A robot that cannot tell when its own perception is unreliable is a hazard. We argue for a self-trust layer that reasons about contradictions across sensors — and acts on them.
Cybersecurity
Hand-written security rules go stale the moment a deployment changes. We report early lessons from building a security baseline that learns itself — and the poisoning traps we are designing against.
AI
Keyword spotting on a microcontroller used to justify throwing the model away. Our 4.2-bit BERT reaches 94.1% accuracy inside 256 KB of RAM. Here is what the accuracy-vs-bits curve actually looks like.
AI
Differential privacy gives federated learning a formal guarantee — and a brutal trade-off screen. We walk through what DP budgets mean when twelve simulated hospitals train one medical imaging model.
Intelligent systems, deep learning, NLP, and computer vision research.
4 open problemsThreat detection, secure systems, cryptography, and privacy engineering.
3 open problemsAutonomous navigation, manipulation, and human-robot interaction.
2 open problemsSelf-driving, drone systems, and multi-agent coordination.
2 open problemsOperating systems, distributed computing, and performance optimization.
2 open problemsData pipelines, analytics, visualization, and statistical modeling.
1 open problemEdge computing, sensor networks, and embedded intelligence.
1 open problemAlignment, robustness, interpretability, and value learning.
2 open problemsReal-time anomaly detection library for autonomous vehicle CAN bus networks. Written in Rust for zero-cost abstractions and deterministic performance.
Muhammad Qasim, Student Team A
Federated learning framework with differential privacy guarantees for medical imaging. Built on Flower with formal privacy accounting.
Student Team B, Muhammad Qasim
A domain-specific language for programming quantized neural networks on microcontrollers. Compiles to TFLite and ONNX.
Student Team C
Student placements, mentored sprints and faculty co-supervision — every program is open to remote participants from any university.
See the programsThree hiring tracks — student researcher, fellow and research engineer — all starting on the same problem board.
Explore careersThe datasets, harnesses and published code the lab built while working its problems — free to use, cite and extend.
Browse resourcesJoin a supervised team on one of the open problems above. A placement is scoped to a single research question, a named supervisor and a deliverable you can put your name on.
Fellowships fund a longer, self-directed piece of work on a problem from the board, with a research mentor and a publication plan agreed at the start.
Fund a shared problem with us through a research grant, or co-supervise a student project with a university group. Existing lab infrastructure is reused rather than rebuilt.
How the lab selects problems and handles data, human participants, authorship, AI risk and misconduct.
Read policyHow to report a vulnerability, what is in scope, what to include and how the lab responds.
Read policyThe standard the lab site targets, what we have done to meet it, known gaps and how to report a barrier.
Read policy