Mahdi Kazemi
Project Technical Lead
SensorGuardian
Building the next diagnostic primitive for ROS2 systems.
A small, selective engineering group tackling real-world sensor failure modes before they become production incidents.
Contributors join as Project Leads and Founding Contributors on a public ROS2 diagnostics initiative.
SensorGuardian is a public ROS2 diagnostics initiative building a lightweight analyzer for early sensor and driver failure detection.
We review applications weekly. Roles are limited to maintain technical quality.
Problem
Robots fail in boring ways first. Before you face deeper issues, you often see frozen drivers, NaNs, and silent degradation. ROS2 Diagnostics catches some issues, but more analysis can reveal more insights.
Solution
SensorGuardian: a public ROS2 diagnostics initiative building a lightweight analyzer for early sensor and driver failures.
Project Leadership & Review
Directing and reviewing the SensorGuardian program.
Saeed Sarfarazi
Project Operational Lead
Behnam Moradi
Project Reviewer and Mentor
Vikrant Shah
Project Reviewer and Mentor
What is it?
A package with a C++ core and a Python API, offering:
- A CLI for easy use.
- An optional ROS2 node for live diagnostics.
In v1, this is basically porting diagnostic_analysis to ROS2 with add-ons:
- Data trend analysis.
- Error logics.
- User-defined error severity and levels.
DATA IN (ROS bag/messages) → (FAST/SIMPLE) ANALYSIS → DATA OUT (To be defined)
What is it not?
- No diagnosis of root causes or fixing sensors.
- No sensor health AI.
Timeline
12 weeks — from prior art review to the first steady and tested version (v1.0).
Not Ready to Commit? Start Small.
- Review and summarize failure modes from one ROS2 bag.
- Draft a minimal CLI UX proposal (1–2 pages).
- Port one diagnostic rule from diagnostic_analysis to ROS2.
- Write a “good first issue” test harness for a target sensor input.
- Map one failure mode to expected WARN/ERROR semantics.
Completing a micro-project is the fastest path into a lead role.
Why contribute?
Contributors will:
- Build public, attributable open-source contributions in ROS2 diagnostics.
- Get real ROS2 systems exposure focused on production-relevant failure modes.
- Gain architecture and API design experience for developer-facing tooling.
- Collaborate directly with Physical AI practitioners and robotics engineers.
- Expand your network through a small, selective contributor cohort.
- Explore new ideas and move toward the startup path when relevant.
Founding Contributor Track (First 12 Weeks)
Limited to 5–7 contributors.
Names listed permanently in:
- GitHub README.
- Project website.
- v1.0 release notes.
Founding contributors receive:
- Verified LinkedIn recommendation.
- Public attribution for architectural decisions.
- Priority access to future CPAI projects.
Open roles
1) Founding Architect — SensorGuardian Core & APIs
You will own the architectural direction of SensorGuardian’s public interfaces and data semantics.
Defines the structure of the system and coordinates team efforts.
Responsibilities
- Define clear, consistent abstractions for validating and consuming data.
- Shape the project’s public APIs and data-contract semantics.
- Balance correctness, usability, and performance in core design decisions.
- Ensure the core remains modular, testable, and extensible as the project evolves.
- Collaborate closely with ROS integration and testing contributors.
- Review design changes to prevent scope creep and architectural drift.
Skills & Requirements
- Strong in C++ (C++17+) and Python.
- Experience with ROS2.
- Experience with high-rate / large-payload sensor data.
- Experience API design and error-handling instincts.
- Ability to reason about real-time constraints and failure modes.
- Knowledge of robotics preferred.
2) Founding Integration Lead — ROS2 & Middleware
You will own how SensorGuardian integrates with ROS2 communication patterns and middleware behavior.
Owns how the project integrates with ROS2 and its middleware ecosystem.
Responsibilities
- Integrate project abstractions with ROS2 communication patterns.
- Ensure correct interaction with ROS2 subscriptions, QoS, and executors.
- Evaluate performance and timing behavior under realistic workloads.
- Provide guidance on ROS2-specific configuration and limitations.
- Ensure compatibility across commonly used ROS2 distributions.
Skills & Requirements
- Strong in C++ (C++17+) and Python.
- Strong experience with ROS2 middleware and DDS concepts.
- Familiarity with ROS diagnostics tooling is a plus.
- Ability to reason about behavior across different ROS2 distros.
- Knowledge of robotics preferred.
3) Founding Quality & Adoption Lead — Testing, QA, Docs
You will own the testing strategy, quality bar, and contributor onboarding experience.
Focuses on testing, documentation, and contributor onboarding.
Responsibilities
- Define and maintain a testing strategy that covers common failure modes.
- Build confidence through automated tests and reproducible examples.
- Maintain CI pipelines and quality gates.
- Write clear documentation explaining when and how to use the project.
- Create onboarding material and good first issues for new contributors.
- Ensure changes are easy to validate and regressions are caught early.
- Advocate for usability and clarity from a contributor’s perspective.
Skills & Requirements
- Strong software testing mindset.
- Strong in C++ (C++17+) and Python.
- Experience with CMake, colcon/amend, and ROS2 build/test workflows.
- Familiarity with CI systems (GitHub Actions).
- Ability to write clear, concise technical documentation.
- Understanding of robotics development workflows.
- Empathy for users and contributors (developer experience focus).
What You Gain by Participating
-
Collaboration Certificate & Industry Referral Signal
Participants who demonstrate strong analytical thinking, reliability, and professional conduct will receive an official Certificate of Collaboration from the program. This certificate is not a generic attendance badge; it documents your contribution to real industry discovery work.
In addition, top contributors may be personally introduced to robotics and Physical AI companies in Canada when relevant roles, collaborations, or project-based opportunities arise. While this is not a job guarantee, it acts as a strong credibility signal and significantly lowers the barrier to serious conversations with hiring managers and founders.
-
Direct Exposure to Real Industry Problems
Instead of relying on papers, courses, or social media hype, you will speak directly with people who deployed Physical AI systems or dealt with their real-world failures. This builds real market intuition—something no classroom can teach.
-
Founder- and Industry-Grade Thinking
Your perspective will shift from “What can we build?” to “What will actually be bought, deployed, and scaled?” This mindset is critical for startups, product roles, and applied engineering careers.
-
Practical Market Discovery Skills
You will learn how to ask high-quality, unbiased questions, separate real pain from surface complaints, and turn ambiguity into clear, actionable insight. These skills transfer directly to startups, product management, VC and investment, and strategy roles.
-
Hands-On Industry Interviewing
You will practice speaking with senior engineers and managers, leading conversations toward operational truth, and identifying bias and hidden constraints. This is a rare and valuable professional skill.
-
Clear Thinking, Distillation, and Writing
You will learn to convert long conversations into sharp insights, write clearly without buzzwords, and define problems crisply and decisively. This skill distinguishes effective founders and engineers.
-
Strong Professional Signal
Participation shows that you have worked on real, ambiguous problems and are market-aware—not just technically skilled. This signal is highly credible to accelerators, early-stage startups, and industrial teams.
-
Public Recognition as a Contributor
With your consent, your name and professional details will appear as a Contributor, published in a public, open-source report, increasing your visibility in Canada’s Physical AI ecosystem.
-
Discover Real Startup Ideas (Not Guesswork)
If you are even slightly interested in founding a startup, this program creates a pipeline of market-validated ideas. Many pain points are pre-idea opportunities. You extract ideas directly from the market.
-
Access to Emerging Physical AI Teams
By participating, you enter the core collaboration layer of the Physical AI Canada community, giving you early access to forming teams, collaboration based on real problems and real skills, and a faster path to building a strong startup team.
SensorGuardian is forming its founding contributor group now. If you want your work to matter — and be visible — apply below.