Norbert Klockiewicz
@Nklockiewicz
Followers
80
Following
117
Media
5
Statuses
56
Mobile ML Engineer – on-device inference (React Native, ExecuTorch) | Open Source Contributor @swmansion
Kraków
Joined February 2025
On a plane? In the woods? Offline? 🤔 Check out new Private Mind – your fast, secure & totally free on-device AI app. 🚀 ✅Chat with your PDFs – Upload documents and ask questions, get summaries, or extract insights. All processing happens locally on your device. ✅ Keep your
3
14
77
More and more people are turning to edge AI! In just last week we had >50% increase in npm downloads! This happened soon after we reached 1k github stars 🌟milestone! And we still have more great things to come!
1
4
17
We just hit 1000 stars on react-native-executorch! When we started this project nearly a year ago, I didn't imagine we'd get here so fast. Grateful for every contributor, issue reporter, and user who's helped shape this. The best is yet to come.
1
3
29
just got my hands on the new iPhone 17 Pro and the perf is insane 90 tokens/s using quantized Qwen 0.6B with react-native-executorch more performance updates on the way 👀
1
2
14
I wrote a blog post for Expo!!!
🔥The future of AI apps is on the device. With react-native-executorch (from @swmansion), you can run models locally in your Expo app: ♢ Private by design ♢ Zero API fees ♢ Works offline ♢ Instant response Read how it works in the blog below from @_darthez_ and
0
1
10
Not only is it faster, but also uses less energy! Faster AND more efficient? YES 🤯! Reaching 91 tokens/sec on iPhone 16 pro for Qwen3 0.6b 🚀 Check out latest version 0.5.8 here https://t.co/14jn43sJjR
Aaaand it's live! Make sure to check it out 🤩
1
4
10
Aaaand it's live! Make sure to check it out 🤩
2x faster LLM inference coming to ExecuTorch soon. Went from 25 tok/s to 62 tok/s on the same prompts. Patch for react-native-executorch drops in a few days 👀
0
1
6
2x faster LLM inference coming to ExecuTorch soon. Went from 25 tok/s to 62 tok/s on the same prompts. Patch for react-native-executorch drops in a few days 👀
0
4
24
Join us for a live stream with @czernekm & @Nklockiewicz this Wednesday at 5:00 PM CEST 🚀 We’ll cover ✅ Pros and cons of on-device AI ✅ Using React Native ExecuTorch in projects ✅ Where to get language models and how to swap them for existing use cases ✅ RAG in React Native
0
6
24
We don’t have flying cars in 2025 yet, but we do have on-device Stable Diffusion in React Native! 👀 Coming soon to React Native ExecuTorch 🚀
3
13
144
.@iwoplaza invites you 🔥to his talk: "TypeScript on the GPU – Bridging the Gap Between Realms." Can’t wait to see you at JSNation US, Nov. 17 & 20!
2
14
91
LLMs are great with language, but not always with facts. They hit cutoffs, miss context, and sometimes just make things up. 🤖 Retrieval-Augmented Generation helps with that by pulling in relevant data. In our new post we broke down how it works, and where it helps (and hurts).
1
7
36
Building new feature for Private Mind app. In the next release you will be able to provide text documents as a context for LLM. React Native ExecuTorch backed up by React Native RAG and Expo gives a powerful combination for building offline AI apps.
1
4
17
Lots of hardwork on improving the internals of library which should allow us to ship features faster 🚀
Running AI in React Native apps just got a lot more stable and faster! ✨ React Native ExecuTorch 0.5.0 brings: ✅ Up to 2x faster Whisper performance on iOS ✅ C++ JSI bindings for zero-copy data transfer ✅ LLM structured output using Zod schemas ✅ CLIP model for image →
0
1
5
Hi everyone, Today we launched Private Mind on Product Hunt. If you are looking for a way to work with local LLMs on your phone you are in right place. Come take a look an leave an upvote! https://t.co/LwWaGudiq1
producthunt.com
Private Mind represents a new era of AI—powerful, personal, and completely offline. Built around the belief that AI should live entirely on your device, Private Mind opens the door to a new kind of...
0
0
2
Your React Native app + React Native Executorch + a few lines of code = LLMs up and running in seconds. 🪄 Check out the docs and get started 👉 https://t.co/QeuQWguncw
1
10
85
Opublikowaliśmy ulubiony dokument wszystkich giełdowych świrów – czyli raport za kolejny kwartał. 💰 Przychody netto ze sprzedaży rosną proporcjonalnie do wzrostu zespołu, czyli o ok. 25%, więc jest się z czego cieszyć. ❤️ A szczegółowe info znajdziecie w wątku poniżej.
2
1
36
Introducing Private Mind! 🚀 Our new open-source app, built with React Native ExecuTorch, lets you run LLMs 100% locally on your device. 🔹 Explore and swap pre-exported models 🔹 Benchmark performance, speed & memory 🔹 Customize system prompts 🔹 100% offline, private & free
6
26
270
Introducing React Native RAG 🚀 On-device Retrieval-Augmented Generation in React Native apps! 🔒 Private: Data stays on-device 📲 Offline: Works anywhere, no cloud needed 🔌 Modular: Integrate external LLMs or your own knowledge base 🔧 Extensible: Seamless with React Native
blog.swmansion.com
We’re excited to introduce React Native RAG. A new library designed to enhance the LLMs capabilities with on-device RAG.
9
49
359