Available for Sponsored Roles · UK / EU / USA

Munawar
Kazmi

Robotics Systems Engineer

Robotics Systems Engineer specialising in Human-Robot Interaction, ROS2, and Edge AI deployment. Delivered a 30% trust improvement on the iCub humanoid through a full sim-to-real pipeline. Open to sponsored roles in UK, EU, and USA.

ROS2 · C++20 iCub Humanoid Edge AI · TensorRT MSc Commendation · UH Isaac Sim · MuJoCo PyTorch · TensorFlow
↗ Get In Touch ↓ Download CV
Munawar Kazmi
0
%
Trust Lift
0
+
Trials Run
0
%
Edge AI Acc.
About Me

Engineer. Researcher. Builder.

I'm a robotics engineer who works at the intersection of human-robot interaction, real-time AI systems, and embedded hardware. My work spans the full stack: from bare-metal C++ firmware on ESP32 microcontrollers to running sim-to-real pipelines on the iCub humanoid platform at the University of Hertfordshire Robot House.

My MSc thesis produced one of the few quantitative studies on humanoid robot companionship, achieving a validated 30% improvement in human trust using a novel blend of verbal and non-verbal communication strategies. The work ran across two humanoid platforms (iCub and JD EZ-Robot) with 200+ structured trials, supervised by Dr Patrick Holthaus.

Beyond research, I've deployed production AI on constrained hardware, cutting CNN inference latency by 40% on Jetson Nano via TensorRT, and engineering deterministic firmware for safety-critical real-world deployments. I care about systems that actually work in the field, not just in simulation.

// Open to Work
Actively seeking Robotics and AI Systems roles in the UK, EU, or USA.
UK Graduate Visa (PSW) valid until September 2026 · Immediate availability.
MSc Artificial Intelligence & Robotics
University of Hertfordshire · Hatfield, UK
Jan 2023 – Aug 2024
Commendation
BEng Mechatronics Engineering
NUST · Islamabad, Pakistan
Sep 2018 – May 2022
Engineering Projects

Built. Deployed. Measured.

Every project ships with real hardware validation and quantified outcomes.

🤖
ROS2 Autonomous Navigation & Deterministic Planning
Developed a high-performance C++20 Global Planner plugin for ROS2 Nav2 using A* and D* Lite. Achieved 22% faster replanning and 24% shorter travel times in dynamic environments, validated on TurtleBot3 with strict protective-stop safety guarantees.
22% faster replanning 24% travel time reduction
C++20 ROS2 Nav2 A* / D* Lite TurtleBot3
View on GitHub →
⚡
Edge AI Optimisation on Jetson Nano
Optimised CNN and gesture recognition models using TensorRT on Jetson Nano. Achieved 40% latency reduction from baseline PyTorch inference while maintaining 96% accuracy, enabling real-time decentralised motor control across a multi-robot system.
40% latency reduction 96% accuracy <200ms inference
TensorRT Jetson Nano PyTorch CNN
View on GitHub →
🛡️
ROS2 LLM Safety Verifier
Built a real-time LLM hallucination detector for ROS2 Nav2 that intercepts unsafe trajectory commands before they reach robot hardware. Validated on TurtleBot3 with Jetson Orin Nano, catching 94% of unsafe trajectories at under 50ms latency across 103 trials.
94% unsafe trajectory catch <50ms on Orin Nano
ROS2 LLM Docker Safety Systems
View on GitHub →
📡
ESP32-CAM Motion Detector (Deployed Product)
Engineered deterministic C++ firmware for image processing on ESP32-CAM. Reduced peak RAM by 60% and built a LoRa mesh protocol achieving under 3s alert latency across 100m field trials. Shipped as a deployed product with 93.8% detection accuracy.
93.8% accuracy 60% RAM reduction <3s alert latency
C++ ESP32 LoRa Mesh Embedded
View on GitHub → ▶ Watch Demo →
🐾
Unitree Go1 Quadruped: LLM Gait Control
Fine-tuned TensorFlow LLMs for robotics inference on the Unitree Go1 quadruped and integrated models directly into the control loop. Achieved 29% latency reduction and 35% better platform responsiveness with 99.5% runtime stability across 200 trials.
29% latency reduction 35% responsiveness gain 99.5% stability
TensorFlow Unitree Go1 LLM Fine-tuning Gazebo
View on GitHub →
🎓
AI & Robotics Curriculum, Pak-Turk Maarif
Designed and delivered an ML curriculum covering neural networks, gesture control, and embedded AI for 30+ students. Teams built working classifiers from scratch, achieving 92% model accuracy and 28% average error reduction on custom classification tasks.
92% model accuracy 28% error reduction 30+ students
ML Education PyTorch Curriculum Design
Read on Medium →
Technical Stack

Tools I Ship With

Robotics & Software
C++20 Python ROS2 (Nav2) ROS Control Plugin Development Git CI/CD Docker
Simulation & Digital Twins
Isaac Sim MuJoCo Gazebo Unity Sim-to-Real Transfer Digital Twin Design
AI & Perception
PyTorch TensorFlow TensorRT Computer Vision Imitation Learning VLA Models LLM Integration CNN Optimisation
Hardware & Embedded
iCub Humanoid Unitree Go1 Jetson Nano ESP32 LoRa Mesh Bare-Metal C/C++ Arduino Safety-Critical Systems
Experience

Where I've Shipped

Visiting Lecturer, AI & Robotics
Aug 2025 – Present
Pak-Turk Maarif Schools · Multan, Pakistan
  • Designed ML curriculum for 30+ students; teams achieved 92% model accuracy with 28% average error reduction on custom classification tasks.
Embedded Systems Engineer
Jan 2025 – May 2025
Muxtronics (Contract) · Multan, Pakistan
  • Engineered deterministic C++ firmware for real-time image processing on ESP32; reducing peak RAM by 60%.
  • Architected a LoRa mesh protocol achieving <3s alert latency across 100m field trials in a safety-critical deployed product.
Robotics Integration Specialist
Jan 2024 – May 2024
Robot House · University of Hertfordshire · Hatfield, UK
  • Architected a gesture recognition and motion planning pipeline in Unity/C++ for the iCub humanoid, reducing user frustration by 40% and improving HRI task efficiency by 25%.
  • Deployed PyTorch-based non-verbal communication models onto iCub's real-time control loop, lifting user engagement from 4.7 to 5.2 on a Likert 1–7 scale across structured interaction trials.
Research Associate
May 2023 – Dec 2023
University of Hertfordshire · Hatfield, UK
  • Fine-tuned TensorFlow LLMs for real-time robotics inference on the Unitree Go1, cutting inference latency by 29% across 200 trials.
  • Integrated optimised models into the control loop, improving platform responsiveness by 35% with 99.5% runtime stability.
Embedded Systems Intern
Mar 2022 – Apr 2022
NCRA, National Centre of Robotics & Automation · Islamabad, Pakistan
  • Optimised bare-metal C firmware for embedded controllers, achieving 40% faster processing and 98% reliability across 100+ safety-critical test cycles.
Publications

Research Output

Enhancing Human-Robot Companionship: Comparing Verbal and Non-Verbal Communication Cues in Humanoid Robots
Kazmi, M. & Holthaus, P. (2025)
Preprint · Target: IEEE RO-MAN 2026
Under Review ↗ Read Paper
Speaking & Teaching

In the Room

Keynote at University of Hertfordshire
KEYNOTE · UNIVERSITY OF HERTFORDSHIRE
AI & Robotics in Education Conference
Delivered a keynote on HRI and Edge AI applications to 50+ students and faculty, with live robot demonstrations of the iCub companionship system.
Teaching AI & Robotics session
WORKSHOP · PAK-TURK MAARIF SCHOOLS
Hands-On Robotics & ML Workshop
Led practical sessions covering gesture control, neural networks, and real-time AI deployment. Teams built and trained working classifiers during the session.
Testimonials

What They Say

"
Munawar demonstrated remarkable technical proficiency, creativity, and a strong work ethic. He actively sought feedback and incorporated it into his work, demonstrating a commitment to continuous improvement. His innovative approach, coupled with a strong foundation in research and development, positions him as a valuable asset to academic or professional institutions. I wholeheartedly recommend him for any future academic or professional opportunities.
Dr Patrick Holthaus
Senior Research Fellow & Visiting Lecturer · University of Hertfordshire · MSc Supervisor
"
Munawar demonstrated exceptional skill in embedded AI deployment during his contract at Muxtronics, leading the ESP32-CAM prototype to 93.8% accuracy and optimizing real-time processing by 60%, a true asset for any robotics team.
Waheed Shah
CEO · Muxtronics
Available for immediate start

Let's Build
Something Real

Open to sponsored robotics and AI engineering roles in the UK, EU, and USA.

🇬🇧 UK Graduate Visa (PSW) valid until September 2026 · Visa sponsorship welcome
✉ Email in LinkedIn </> GitHub ✍ Medium ↓ Download CV
munawarsaeedkazmi1@gmail.com