About
Shubham Malhotra is co-founder and CTO of RunAnywhere. He previously worked on EC2 Spot at Amazon and Azure at Microsoft, and has authored peer-reviewed research. He developed MetalRT, a multi-modal inference engine for Apple Silicon with custom GPU kernels for on-device voice AI processing.
Shubham Malhotra @shubhammal72313
"free. Private. Fast. Controllable." Clem Delangue summed up local AI better than most of us could. The models are getting good enough. Now the hard part is making them easy to build with and easy to ship. That's exactly what we're trying to solve at @RunAnywhereAI. https://t.co/G78WD0IfbA
Jul 31, 2026 · ♥ 3 · ⇄ 2
Shubham Malhotra @shubhammal72313
Check out the RunAnywhere SDKs: https://t.co/MPijjsrzI6
Jul 29, 2026 · ♥ 2
Shubham Malhotra @shubhammal72313
RAG doesn’t need a cloud vector database. This demo uses @RunAnywhereAI to chunk a PDF, generate 384D MiniLM embeddings, run local cosine search in 3.4 ms, and generate cited answers with Qwen—fully on-device. No Pinecone. No API calls. SDK link in replies. https://t.co/6ILLOpYotx
Jul 29, 2026 · ♥ 8 · ⇄ 2
Founder at RunAnywhereYC W26
Intro paths
Shubham went through Y Combinator (Winter 2026) — any YC alum in your network can reach them through the YC community.
Their group partner at YC was Diana Hu, General Partner at Y Combinator — the partner who ran RunAnywhere's batch and knows the company first-hand. YC profile ↗
In turn, Shubham can intro founders to RunAnywhere's investors below.
