ChipJev released: AI-powered analog circuit design with open-source EDA
ChipJev combines fast AI decisions with GPU-accelerated topology and sizing search. Try the live SKY130 circuit design demo at chipjev.com and explore the open-source code.
TU Delft · Department of Microelectronics
Intelligence.
Made efficient.
We design algorithms and hardware together to bring energy-efficient intelligence to the edge.
Lab of Efficient Machine Intelligence
Delft, the NetherlandsAlgorithms ↔ Hardware
Inside EMI
Latest from the lab
News & perspectives.
GrainSpeech preprint: compact speech synthesis with finer detail
Our new GrainSpeech arXiv preprint introduces a 264.8K-parameter acoustic model for compact speech synthesis. The code, pretrained model and audio demos are available online.
MSc Navya Agarwal’s Work Accepted at ICCAD 2026
Congratulations to EMI MSc student Navya Agarwal, whose internship research at imec has been accepted for presentation and publication at the 2026 IEEE/ACM International Conference on Computer-Aided Design (ICCAD 2026).
Our research
From learning algorithms
to efficient silicon.
Our work connects neuromorphic computing, signal processing and hardware design.
01
AI for RF & mixed-signal systems
Learning-based models, calibration and digital predistortion for more efficient radio-frequency and mixed-signal systems.
Digital predistortionPhoto: OpenDPDv2 · Fig. 302
Sparse neural computing
Brain-inspired algorithms and accelerators that exploit temporal and spatial sparsity to reduce unnecessary computation.
Dynamic sparsityPhoto: Spartus · Fig. 1003
Event-based vision & wearable intelligence
Efficient visual perception, eye tracking and intelligent hardware for extended reality and wearable applications.
Event camerasVideo: Q. Chen & C. Gao · 202304
Speech & intelligent sensing
Energy-efficient speech processing, keyword spotting and radar-based activity recognition for embedded systems.
Speech processingPhoto: DeltaKWS · Fig. 9(a)Explore our open-source research.
Find code, datasets and pretrained models from EMI on GitHub.
Selected publications
Ideas worth
building on.
Explore our work in efficient AI, circuits, systems and sensing.
Browse publicationsOpenDPDv2: A Unified Learning and Optimization Framework for Neural Network Digital Predistortion
Y. Wu, A. Li, C. Gao
GLOBECOM · AcceptedBitFair: A 12nm Bit-Serial CNN Accelerator with Learnable Early Termination and Adaptive Bit Ordering for Ultra-Low-Power XR Vision
A. Li and C. Gao
IEEE JETCASExploiting neuro-inspired dynamic sparsity for energy-efficient intelligent perception
S. Zhou, C. Gao, T. Delbruck, M. Verhelst, S.-C. Liu
Nature Communications
People behind the research
Good science.
Shared curiosity.
Led by Dr. Chang Gao, we are a team of researchers and students at TU Delft, working across machine learning, microelectronics and embedded systems.
Get to know us
Supported by
Build with us
Curious about what comes next?
Explore MSc projects, research opportunities and collaborations.
Around the world
Our global visitors.
—recorded visitors
—countries & territories
Loading visitor statistics…
Select a country on the map or in the list.
| Country / territory | Visitors |
|---|
Loading country data…
Source: Google Analytics
About these numbers
Historical visitors are distinct users recorded by Google Analytics through yesterday. A person can appear in more than one country, so country counts may not add up to the total. The recent view shows users active within the last 30 minutes and refreshes every minute while this section is visible. Locations are approximate; unreported or unavailable data is not counted as zero. Analytics may omit visits or withhold some data.