We are thrilled to feature in Tech AI Magazine's list of 2026 AI Companies! 🎊 It is so great to see energy innovation highlighted on an editorial dedicated to covering the industry’s most relevant tech news. Check out their publication and read the profile on Mythic here 👉 https://lnkd.in/gDbTe-uj
Mythic
Semiconductor Manufacturing
Austin, TX 16,001 followers
Pioneering Analog Compute for AI
About us
Founded in 2012 by Mike Henry and Dave Fick, and based in Austin, TX and Redwood City, CA, Mythic is creating a unified hardware and software platform that relies on unique analog compute-in-memory technology to deliver revolutionary power, cost, and performance that will shatter the limits restricting AI innovation. Mythic is making it much easier and more affordable to deploy powerful AI solutions, from the data center to the edge device.
- Website
-
http://mythic.ai
External link for Mythic
- Industry
- Semiconductor Manufacturing
- Company size
- 11-50 employees
- Headquarters
- Austin, TX
- Type
- Privately Held
- Founded
- 2012
- Specialties
- Artificial Intelligence, Machine Learning, Computer Vision, and Neural Networks
Locations
-
Primary
Get directions
Austin, TX, US
-
Get directions
Vancouver, CA
-
Get directions
Silicon Valley, US
Employees at Mythic
Updates
-
Mythic reposted this
In a world of intelligence inevitability, how we get there is a choice. At Mythic, we believe growing intelligence shouldn't cost us the planet. While others race toward the next frontier of human civilization, we're focused on making sure our planet gets there with us. Come build with us. → https://mythic.ai/join-us/
Honda Xcelerator Ventures portfolio company Mythic made a big impact at SXSW with CEO Taner Ozcelik’s keynote on “The AI Energy Crisis” hosted by the Team RTX. It’s one of the most important conversations happening in AI today. This challenge spans: • AI infrastructure • Edge computing • Robotics & autonomous systems To put things in perspective, Taner highlighted the scale of the human brain: ~86 billion neurons and ~100 trillion synapses, all with ~20 watts of power, while the AI datacenters are at megawatt scale. The next phase of this market isn’t just about better chips or more efficient AI models—it’s about energy-aware systems, distributed compute and energy, and solving power delivery at the last meter. Mythic is leading the way with AI inference at the silicon level, a game-changer for powering the next era of AI innovation. Learn more: https://lnkd.in/gtrXVEy2 #HondXceleratorVentures #HXV #SXSW #Mythic
-
-
Nothing beats The Bay... or Mythic's 100x energy advantage! 🌁 Join our CEO, Taner Ozcelik, right here in San Francisco on Wednesday at 10:20 for Building for AI Scale: Inside the Next-Gen Infra Stack 📍 #HumanX2026 #AnalogProcessingUnit 🧣 https://lnkd.in/gUhZYbxA
-
-
A timely conversation for analog processing indeed ⏰
We're heading to Austin this weekend for SXSW! ✈️ Find out why everyone is talking about analog computing from an ex-NVIDIA founder. This Sunday at noon, March 15th, join our CEO Taner Ozcelik for this rare opportunity. Secure your spot: https://luma.com/u4q9akh7
-
-
Mythic reposted this
AI Chips Just Got 100x More Efficient?! Mythic’s Ultra-Low Power Breakthrough Explained 🚀 What if AI could run 100x more efficiently while using ultra-low power? 🤯 In this video, we explore how Mythic is teaming up with Microchip Technology to revolutionize AI hardware using memBrain and SuperFlash memory. With up to 120 TOPS per watt, this new analog compute-in-memory technology could transform edge devices, data centers, and the future of AI processing. ⚡ 👉 Discover how this breakthrough could change everything from smartphones to autonomous systems. 📌 Don’t forget to like, share & subscribe for more AI tech updates! Read more: https://lnkd.in/gYPKR_Ak #AI #ArtificialIntelligence #TechNews #AIChips #Semiconductors #EdgeAI #MachineLearning #DeepLearning #FutureTech #Innovation #DataCenters #ChipDesign #Electronics #Startup #TechTrends #ViralTech
-
Translation available for our supporters in Taiwan! 👋 https://mchp.us/4lAlEsU
Mythic新一代APU採用SST的memBrain™記憶體單元，具備多位元儲存能力、奈安培等級讀取電流、長時間資料保存，以及單週期MAC運算能力，使類比AI推論效能可達每瓦120 TOPS。 這項技術為邊緣AI及更多AI應用帶來突破性的能源效率。了解更多︰https://mchp.us/4lAlEsU #MythicAI #AIInference #AI推論 #能源效率運算
-
-
Hi Japan! ✍ https://mchp.us/4lzLiOB
【ニューロモルフィック メモリでエッジAIの電力問題を解決】 AIの利用が急速に広がる中、その電力消費の大きさが課題となっています。この課題を解決する鍵として注目されているのが、Microchip社の子会社SST社のmemBrain™技術です。Mythic社の次世代APU(アナログプロセッシングユニット)にも採用されたmemBrain™ニューロモルフィック メモリは、マルチビット ストレージ、1桁nA台の読み出し電流、長期のデータ保持、シングルサイクルのMAC(積和)演算を特長とし、120 TOPS/ワットという画期的な電力効率を実現します。詳細：https://mchp.us/4lzLiOB #MythicAI #AI推論 #低消費電力コンピューティング
-
-
We like our partnerships the way we like our four-leaf clovers: handpicked. 🍀 Alongside our work to co-develop the next generation of ultra-low-power vehicles with Honda, we’ve selected Microchip Technology Inc.'s neuromorphic memBrain™ technology for our latest Analog Processing Units. Utilizing SST’s SuperFlash® memory, with 150B units shipped to date, Mythic’s #APUs achieve 120 TOPS/watt with 8-bit precision. https://mchp.us/4uwoHXi
-
We're heading to Austin this weekend for SXSW! ✈️ Find out why everyone is talking about analog computing from an ex-NVIDIA founder. This Sunday at noon, March 15th, join our CEO Taner Ozcelik for this rare opportunity. Secure your spot: https://luma.com/u4q9akh7
-
-
Mythic reposted this
We’re scaling Mythic — and we’re hiring two pivotal areas that will define how developers experience our hardware and how fast it runs in the real world. After an oversubscribed $125M round and a new joint development partnership with Honda, we're growing rapidly to deploy analog compute-in-memory to vehicles, datacenters, and robotics. At the core is our Dataflow Processing Unit (DPU): a graph‑first digital architecture where the ISA describes nodes + edges directly — purpose‑built for neural nets and streaming/graph workloads. We’re hiring key leaders in: COMPILER ENGINEERING Help turn graphs into blisteringly efficient execution on a novel dataflow architecture. • MLIR / PyTorch Infrastructure: https://lnkd.in/gBrHfzcX • Machine Learning Compiler: https://lnkd.in/gFfYznuc • Algorithmic Workloads Compilation: https://lnkd.in/gjEKMgKn ARCHITECTURAL DESIGN AND MODELING Build the models that guide tradeoffs, de-risk RTL, and enable software *before* silicon exists. • Architecture Modeling Engineer: https://lnkd.in/gNV5Bgw4 • Senior Silicon Emulation Engineer: https://lnkd.in/gbJ8B3MF • Senior RTL Design Engineer: https://lnkd.in/gj5-HHPn This is one of those rare moments to join: 14 years of deep tech development + real momentum + the chance to lead from the front. If this sounds like you (or someone you know), DM me or apply directly. Referrals welcome. 🙏 We have offices in Austin, Palo Alto, and Bangalore, and many people working fully remote as well #hiring #compilers #MLIR #PyTorch #computerarchitecture #SystemC #AIhardware #edgeAI #automotive