Inference Happy Hour for AI Founders & Builders Hosted by Massed Compute, Rapt AI, Amaara, Qwerky AI, & Modular.
Inference workloads are exploding. The question every AI builder and founder is facing isn’t can the model run — it’s can it run fast enough, reliably enough, and profitably enough in production.
Join us for an Inference Happy Hour where AI founders and builders are pushing model limits with new real world discoveries: meet the teams redefining AI performance, efficiency, and scale.
If you're creating or deploying LLMs, running inference, reducing GPU burn, or preparing your infrastructure for the next wave of inference demand — you should be here.
Think It. Build It. Scale It.
Your hosts:
Massed Compute - Frontier AI cloud for inference and training. Bare metal GPUs with up to 70% savings vs. the hyperscalers. NVIDIA Cloud Partner.
Rapt AI - Model-Defined GPU™ optimization that dynamically allocates exactly the resources each model needs in real time—delivering 7–12× more models at your required latency.
Qwerky AI - Optimize large language models for 4x speed and 75% cost reduction — no hardware changes required. Making AI Faster, Cheaper, and more Efficient.
Amaara Networks - A systems integrator and value added reseller of technology solutions.
Modular - High-performance AI inference built on Mojo, an AI-native systems language, providing full-stack control from custom GPU kernels to production cloud endpoints
Details:
🗓️ When: Wednesday, March 18, 2026 · 5:00 – 9:00 PM PT
📍 Where: Downtown San Jose (location shared upon RSVP)
🍸 Format: Evening Dinner & Drinks. Curated guest list. No slides.
🎟️ Cost: Complimentary (invite-only, approval required)
This is a curated, invite-priority networking event — not first-come, first-served. Every registration is reviewed to ensure the room stays focused on serious AI builders and founders. Capacity is limited and expected to fill quickly. If you received this, it’s intentional. A GTC conference pass is not required to attend. 350 GPU hour raffle to end the night.