Robotics and AI engineer building Physical AI — intelligent systems that perceive, reason and interact with the physical world. Co-founder and CTO of RoboRecs.
Software that only has to be right on a benchmark is a different discipline from software that has to be right on hardware, in a room it has never seen, on the first attempt. Physical AI is the second one, and that constraint shapes everything below.
→ Robustness beats benchmark scores. A model that is two percent better on a test set and unpredictable on real hardware is not an improvement. Building AI for a medical device taught me that the interesting failures are the ones that never appear in the validation split.
→ Calibration is not a footnote. Most of the gap between a research result and a working product is measurement: knowing what your sensor actually reports, how it drifts, and what your model quietly assumes about it. That work is unglamorous, and it is where the reliability comes from.
→ Embodiment changes the problem. An agent that can act has to reason about consequences it cannot undo. Perception, reasoning and control stop being separable modules once there is a body in the loop, and designing them as one system is the whole point.
→ Research becomes product on contact with hardware. Moving from a paper to a deployed system is not an implementation detail. It is most of the work, and it is where the real design decisions get made.
→ Build for the environment you will actually deploy in. Complex, unstructured, poorly lit, partially observed. Designing for the clean case and hardening afterwards is a reliable way to build something that never quite works.
Built to adapt, engineered to solve.
Get in touchTwo products in production, in two very different sectors — both about turning a hard technical problem into something people can actually use.
Egocentric, multimodal human demonstration data for Physical AI
Humanoid robot foundation models need millions of hours of manipulation data. Roughly five thousand hours of open data exist. RoboRecs closes that gap with head- and wrist-mounted capture that records tasks from the operator's viewpoint — the same viewpoint the robot will have — across RGB, depth, IMU, audio, pose and force. EU-jurisdiction data, captured under GDPR and the AI Act.
Bills of quantities for the Bulgarian construction sector, in ten minutes
Construction cost estimation in Bulgaria still runs on spreadsheets and guesswork. Buildly generates a complete bill of quantities online from a library of more than 4,000 work items priced at real market rates, then carries it through to quotes and invoices. A different sector from robotics, and the same engineering problem: replacing an expert's slow manual process with something reliable and fast.
I am a robotics and AI engineer with an MSc from EPFL, focused on Physical AI. I build intelligent systems that perceive, reason and interact with the physical world. My experience spans robotics, computer vision, machine learning and real-world system deployment across medical imaging and autonomous systems.
At SamanTree Medical I developed image quality assessment algorithms for histological scanners, working at the intersection of computer vision, machine learning and medical device calibration. Building AI for a medical device taught me the importance of robustness, reliability, and systems that perform consistently outside the lab.
I am currently co-founder and CTO of RoboRecs, building the egocentric human demonstration data that humanoid robot foundation models are starved of. I enjoy problems at the intersection of AI, robotics and real-world deployment, with a focus on turning research into systems that operate reliably in complex environments.
My expertise covers Physical AI, robot perception, computer vision, image processing, deep learning, robotics, Python, C++ and intelligent system design.
Born and raised in Sofia, Bulgaria, with academic and practical experience across Switzerland, Mexico, Washington D.C., The Hague and China. I have taught mathematics, numerical analysis and robotics at EPFL, and ran robotics clubs for children for four consecutive years.
I founded the Association of Bulgarian Students in Lausanne (ABSL) — the first organisation of its kind for Bulgarians in a Swiss academic environment — and served as its president for five consecutive terms, building a community of over 100 members from EPFL, UNIL and HEC Lausanne.
I am always interested in connecting and collaborating with engineers, founders and researchers building the next generation of Physical AI and intelligent robotics.
Trained at institutions with international intake, strict standards, and work that had to actually run.
From international robotics olympiads to diplomatic forums — competing and building at world level.
Engineering competence applied to systems that have to work on real hardware.
Medical imaging, autonomous systems, and the transition from a research result to something that runs reliably.
Engineering solutions to real problems — from medical imaging to autonomous robotics.
You learn as much about an engineer from what they do when nobody is grading it.
Always interested in connecting with engineers, founders and researchers building the next generation of Physical AI and intelligent robotics.