1/2
Meet #NullClaw 🦞, a full AI agent framework written in raw #Zig that fits in 678 KB, runs on ~1 MB of RAM, and boots in under 2ms.
The entire thing is ~45,000 lines of Zig with 3,230+ tests.
🔗
github.com/nullclaw/nul...
#AIAgents #EdgeAI #EmbeddedAI #OpenSource #TinyML #AIInfrastructure #AI
A TinyML Wearable System for Real-Time Cardio-Exercise Tracking
www.mdpi.com/2673-4591/11...
By Timothy Malche
From the 12th International Electronic Conference on Sensors and Applications
#TinyML #WearableTech #FitnessTech #CardioMonitoring
Harvard's open-source ML systems engineering textbook with runnable TinyTorch framework - bridging t
Harvard's open-source ML systems engineering textbook with runnable TinyTorch framework - bridging the gap between AI theory and production-ready systems
https://github.com/harvard-edge/cs249r_book
#MLSystems #TinyML #EdgeComputing
Full object detection on a #microcontroller in 3.5 ms is quite impressive! Yes, I’m using an #AI accelerator…I’m certainly not doing inference on the CPU in that amount of time 😁 #edgeAI #tinyml #embedded
So, #TinyML is too small. LLMs are...probably not what I am looking for.
Is there like a GoldieLocksML or something? #ai #ml #machinelearning
Open source frameworks are driving TinyML to embed powerful machine learning on ultra-low-power microcontrollers, shifting intelligence from the cloud to the smallest connected devices for smarter, faster responses.
www.opensourceforu.com/2026/01/the-...
#electronics #TinyML #opensource
#Science #Technology #AI #Innovation #NASA #SpaceExploration #Biotech #GeneEditing #ClimateAction #Geoengineering #FutureTech #DeepSeek #TinyML #STEM #AgTech #Australia #DigitalPolicy #Research
What if your ESP32 could talk — without the internet?
Check out this offline text-to-speech project using ESP32 that brings voice to your microcontroller projects!
👇 Take a look:
🔗 circuitdigest.com/microcontrol...
#ESP32 #TinyML #TextToSpeech #EmbeddedAI #IoT #EdgeComputing #Maker
The discussion delved into "Minimally Viable LLMs." How small can a language model be and still retain conversational ability? Z80-μLM demonstrates that tiny models can offer surprising capabilities, challenging assumptions about AI scale. #TinyML 3/6
Pushing a $7 Linux Board to the Limit NPU Inference on the LuckFox Pico RV1103 Continue reading on Medium »
#rockchip-rv1103 #npu #machine-learning #tinyml #luckfox-pico
Origin | Interest | Match
ICYMI: I joined John Koetsier on the TechFirst podcast to talk about how quantum-inspired AI compression is reshaping what’s possible for AI at the edge.
Catch the full episode here → podcasts.apple.com/us/podcast/f...
#EdgeAI #AICompression #TinyML #FutureOfAI
News from #STMicro as the company updates its #tinyML model zoo - now some 140-models strong, plus with fresh support for #PyTorch after previously pinning its hopes on #tflite […]
Machine Learning Trends: The Future is Here #explainableAI #AIethics #reinforcementlearning #federatedlearning #AItrends2024 #tinyML #generativeAI #naturallanguageprocessing #MLOps #futureofmachinelearning
Running deep learning models on microcontrollers has gained mainstream popularity among silicon vendors and #firmware developers. My latest article examines the state of the #TinyML runtime offerings in 2025.
👇👇👇
shawnhymel.com/2994/deep-le...
#microcontroller #embedded #AI #MachineLearning
Top Machine Learning Trends to Watch in 2024 #explainableAI #datasciencetrends #tinyML #largelanguagemodels #AutoML #MLOps #AItrends #AIinbusiness #futureofmachinelearning #causalAI
#AllThingsOpen 2025 was incredible! 🙌 Huge thanks to everyone who joined my session, "TinyML Meets PyTorch: Deploying AI at the Edge with Python Using ExecuTorch." The live demo energy was 🔥
Grab the slides + code here 👉 bit.ly/4mNZL8R
#EdgeAI #TinyML
"What skills do I need for edge AI?" is a question I often get. #EdgeAI requires a unique blend of #ML math, #Python for training/optimization, and #embedded or #mobile development. Check out my full blog post to learn more (and recommended courses):
👇
shawnhymel.com/2976/how-to-...
#TinyML
Definitely want to experiment with the #Arduino UNO Q: dual-core Cortex-M7, onboard NPU, and TensorFlow Lite support. Real-time #AI at the edge, no cloud required
🔗 arduino.cc/product-uno-q #Arduino #EdgeAI #TinyML #IoT
LORT Model Boosts Speech Enhancement with Tiny 0.96M Parameter Design
LORT, a compact 0.96 M‑parameter model, delivers strong monaural speech enhancement on VCTK+DEMAND and DNS Challenge datasets while keeping computational load low. Read more: getnews.me/lort-model-boosts-speech... #speechenhancement #lort #tinyml
Dual‑Hand Gesture Interface with TinyML Boosts ROS Mobile Manipulators
Dual‑hand TinyML gesture interface lets one hand steer a robot’s base and the other control a 7‑DOF Kinova arm in real‑time via ROS. Paper posted 23 Sep 2025. Read more: getnews.me/dual-hand-gesture-interf... #tinyml #ros #kinova
TinyAC: Enabling Autonomous Computing for Tiny Edge Devices
TinyAC introduces a two‑layer system: ultra‑light TinyML models run on microcontrollers with just a few kilobytes of RAM, while cloud‑hosted LLMs handle high‑level policy and explanations. Read more: getnews.me/tinyac-enabling-autonomo... #tinyac #tinyml #llm
TinyML benchmarks finally address real-world deployment with MLCommons' new streaming benchmark in MLPerf Tiny v1.3. Tests 20-minute continuous wake word detection while measuring power and duty cycle.
Technical deep dive: mlcommons.org/2025/09/mlpe... #MLPerf #TinyML #EdgeAI
Advancing Stress Detection and Health Monitoring with Deep Learning Approaches
www.mdpi.com/2813-0324/10...
By Merouane Mouadili et al.
From the SCGT’2025 Conference
#TinyML #AIinHealthcare #WearableTech #HRV #DeepLearning
Read the latest work on a novel, computationally efficient training framework with bio-signal processing on edge medical devices: ow.ly/L1bX50WYFEp
By Zhaojing Huang et al. (@sydney.edu.au)
#EdgeAI #TinyML #BioSignalProcessing #MedicalAI #OpenAccess #ModelOptimization
GhostNetV3‑Small Boosts Tiny Image Classification and Distillation Review
GhostNetV3‑Small, a compact network tuned for 32 × 32 images, reaches 93.94% accuracy on CIFAR‑10; the paper was submitted on 15 September 2025. Read more: getnews.me/ghostnetv3-small-boosts-... #ghostnetv3 #distillation #tinyml
The Role of Edge AI and Tiny ML in Modern Robots - buff.ly/eZVIUMV #iot #industrialiot #edgeai #tinyml #robots
![Doug Whitfield [Minneapolis]](https://cdn.bsky.app/img/avatar/plain/did:plc:trn7ilya4fomeojird75epwb/bafkreiaywmnr7s5ms47e5ktmnnzmhhligcaa3hydmyglhpptq5xdbpvype)
