About this Event
Hear 4 expert talks, take the mic for an AI / hardware demo or announcement, and connect over pizza & drinks.
Important: Please grab your here to attend. Registration is required, and tickets will be checked at the door.
A special thanks to the University of Melbourne for hosting us, in collaboration with the Melbourne University Electrical Engineering Club (MUEEC) — a student-run club enriching the student experience through social events, professional development, and industry networking that connects engineering students with peers and professionals.
Pizza and drinks for this session are kindly provided by Advantech — a global leader in IoT intelligent systems and embedded platforms, with the corporate vision of “Enabling an Intelligent Planet.”
Take the mic at our Community Open Mic! Bring your demo, prototype, or announcement — whether it is AI, hardware, robotics, embedded systems, sensing, or edge computing. All welcome.
Agenda
5:30 – 6:00 pm | Check-in, Welcome & Introduction
Hear a short welcome from the event hosts and an introduction to the Physical AI Meetup community.
6:00 – 6:20 pm | Talk 1 –Integrating External Conditioning into Large Pretrained Deep Learning Models
Overview:
Off-the-shelf AI models are impressively capable — until they are used in highly specific real-world domains. A model trained on broad internet-scale data may struggle with cancer biopsy analysis, flood prediction, recognising unfamiliar objects, or following complex multi-step instructions.
In this talk, Dr Sachith Seneviratne argues that the answer is not simply to retrain models from scratch or add more data. Instead, he introduces the idea of external conditioning: injecting domain knowledge into large pretrained models at the right point, with minimal engineering overhead.
Drawing on recent work across medicine, climate, safety-critical AI, and creative tools, Sachith will show how external knowledge can help pretrained models become more reliable, efficient, and useful in real applications. Attendees will leave with practical strategies for making large AI models work in their own engineering and applied AI contexts.world systems.
Speaker:
Dr Sachith Seneviratne, University of Melbourne
Bio:
Dr Sachith Seneviratne is a Research Fellow in AI at the University of Melbourne. He brings over a decade of AI research experience alongside four years of applied consulting work for organisations including NASA and Harvard University.
His research centres on representation learning, applied AI, and the efficiency of deep learning systems. His work asks how large AI models can be made to operate reliably in specific real-world domains — from clinical pathology and flood emergency response to cybersecurity and urban design — without sacrificing the generality that makes them powerful. His research has appeared at leading venues including CVPR, ICCV, and AAAI.
Sachith has developed AI tools for earthquake prediction, malware detection, wearable health monitoring, and automated urban design generation. His PhD from Monash University investigated the automation of machine learning itself, including generating parallelised inference algorithms directly from model specifications.
As a consultant, he has advised global organisations on deploying machine learning in production, with a focus on building systems that are efficient, robust, and grounded in domain knowledge.
6:20 – 6:40 pm | Talk 2 – Harnessing Intelligence from the Source: Living Biological Neurons as a Material for Physical AI
Overview:
What happens when intelligence is no longer built only in silicon?
In this talk, Dr Brett J. Kagan will explore the frontier of biological computing and synthetic biological intelligence, where living neural systems are integrated with hardware and software to create new forms of adaptive computation.
Drawing on Cortical Labs’ work with neurons on a chip, closed-loop learning systems, and the CL-1 platform, Brett will discuss how biological neural systems can learn, adapt, and interact with their environment — and what this could mean for the future of AI, robotics, embodied intelligence, and physical AI.
Speaker:
Dr Brett J. Kagan, Chief Scientific Officer / Chief Operations Officer, Cortical Labs
Bio:
Dr Brett J. Kagan is the Chief Scientific Officer and Chief Operations Officer at Cortical Labs, a multidisciplinary deep-tech startup integrating hardware, software, and synthetic biology to explore how intelligence can be harnessed from neurons on a chip.
Dr Kagan has a PhD in neuroscience focusing on stem cell therapy and completed post-doctoral work in bioinformatics and regenerative medicine. His recent work includes developing the first real-time closed-loop demonstration of in-vitro intelligence in a simplified Pong-game environment, along with work to better test and understand these systems.
He has led the scientific development of the first commercialisable device for harnessing the information processing capabilities of biological neurons, called the CL-1. He also explores the neurocomputational, philosophical, and ethical implications of this emerging technology.
6:40 – 7:00 pm | Pizza & Networking
Enjoy pizza, drinks, and time to connect with other attendees.
7:00 – 7:20 pm | Talk 3 – Physical AI in Practice: Connecting Edge Intelligence to Industrial Outcomes
Overview:
Physical AI becomes truly valuable when intelligence can be deployed close to machines, sensors, factories, and real industrial operations.
In this talk, Susie Chen will share how edge AI, industrial IoT, and data-driven decision-making are being applied in real manufacturing and industrial environments. She will discuss how organisations can connect intelligent edge systems to practical outcomes, including operational efficiency, automation, smarter monitoring, and improved decision-making.
The session will provide an industry perspective on how Physical AI moves from concept to deployment — and how edge intelligence can help bridge the gap between AI innovation and measurable business value.
Speaker:
Chen Susie
Bio:
Susie Chen is the EIoT Head of Sales at Advantech Australia, specialising in edge AI and smart manufacturing. With global experience across Taiwan, Brazil, and Australia, she helps organisations transform industrial operations through AI-driven IoT and data-powered decision-making.
7:20 – 7:40 pm | Talk 4 – Why AGI Needs a Body: The Case for Proprioceptive Data in Embodied Intelligence
Overview:
Vision-based AI has transformed perception, but real-world intelligence requires more than seeing. For robots and embodied systems to operate reliably in the physical world, they need access to ground-truth data about movement, force, position, and physical interaction.
In this talk, Anantyash Dixit will argue that proprioceptive data — the body’s sense of movement and position — is a missing ingredient in the path toward embodied intelligence. He will discuss why specialised sensing hardware is needed to capture this data, how human movement datasets can support robotic deployment, and why physical AI may require a deeper connection between hardware, biomechanics, and intelligence.
Drawing from Melbourne Bionics’ work across wearable sensing, athlete performance analytics, and robotics datasets, this talk will explore how real-world proprioceptive data could shape the next generation of embodied AI systems.
Speaker:
Anantyash Dixit, Founding Engineer, Melbourne Bionics
Bio:
Anantyash Dixit is a Founding Engineer at Melbourne Bionics, a medtech startup where he leads the full hardware pipeline, including electronics design, embedded systems, and sensor integration.
Melbourne Bionics is developing three connected products: Ares, a calf sensor sleeve that captures real-time proprioceptive data from elite athletes; Athena, an analytics platform that turns that data into actionable insights for coaches and medical staff; and Richter, a robotics platform that uses the same dataset as a proprietary benchmark for humanoid robot locomotion.
Anantyash’s background spans FPGA and RTL design, embedded firmware, PCB layout, EMI/EMC verification, wearable biosignals, and medical device development. He holds a degree in Electrical and Computer Systems Engineering from Monash University, where he was awarded the International Excellence Scholarship and served as President of the Society of Monash Electrical Engineers.
Beyond hardware, he thinks deeply about the ethics of AI, the implications of physical intelligence, and what embodied systems mean for humanity’s long-term trajectory.
7:40 – 8:00 pm | Community Open Mic
Take the mic and share with the group — from hardware demos and embedded prototypes to quick AI-at-the-edge announcements, project showcases, calls for collaboration, and more.
Wrap-up & Social
Any further Q&As and chats will happen here.
Last reminder: Please make sure to register on Eventbrite for your free ticket — it is required for entry.
Event Venue & Nearby Stays
Electrical & Electronic Engineering Building (Building 193), 100 Grattan Street, Parkville, Australia
AUD 0.00
