Builder Lunch with Andrew Ng: SideStage
Live commerce sellers run shows where chat, product answers, listings, inventory, and conversion all happen at once, in real time. Operators are understaffed, and the work is messy, high-stakes, and hard to control as a show scales.
We are exploring SideStage, a live commerce operator copilot that turns messy real-time show operations into a controllable, guardrailed workflow. It consumes chat, product, and transaction events on one stream and gives operators a single surface for real-time chat synthesis, guardrailed replies, listing and inventory control, and multilingual host support. The wedge is operator trust during live shows, not generic chatbot quality.
This lunch is an informal working conversation about what it takes to build a real-time operator system that earns trust during live shows: low-latency event handling, guardrailed automation operators can rely on, and the right primitives for high-stakes live commerce.
The Engineer in Residence program is a 12-week Mountain View residency with AI Fund. EIRs work directly with Andrew Ng and the AI Fund team to pressure-test an applied GenAI idea and decide whether to build a company around it. Read the EIR FAQ for more details.
Who's a good fit for this conversation:
Builders who have shipped real-time systems, streaming, or low-latency applications
Engineers comfortable with edge stacks such as Cloudflare Workers, Durable Objects, and server-sent events
Founders or technical leads who have built operator-facing or high-stakes live UX
Applied AI builders who care about guardrailed automation and operator trust
What to expect:
Lunch in Mountain View, CA
Small group, 15-20 builders
Andrew Ng and the AI Fund team
Open technical conversation
Invite-only. Capacity 20.