Founding Engineer (Backend & Infra)

Got strong opinions on how a real-time AI system should be built, and the itch to rebuild what's wrong?

A YC-backed startup is automating the borrower side of US mortgage lending with AI voice and SMS agents. They handle payments, escrow questions, loan qualification and collections for several of the ten largest US mortgage lenders, with every call checked for compliance (FDCPA, RESPA, TILA). $1B+ in payments processed so far, and the team will roughly triple by year-end.

The official title is Member of Technical Staff, but in practice you're a founding engineer, leaning infra and backend. They're looking for the engineer who walks in, forms a view on how the product should really be built, and starts rebuilding it.

They're hiring 4 founding engineers.

What you'll own

• The core systems behind real-time voice AI and compliance workflows
• Keeping latency under 700 ms with event-driven design and parallelism
• Running and scaling fault-tolerant services on Kubernetes
• An AI system that learns from each customer conversation and from human feedback, and keeps getting better
• Taking designs from the whiteboard to production inside banks' own environments
• Setting the engineering standards for the team that's coming

What you bring

• 3+ years of professional engineering, with strong Python fundamentals
• You've designed production backends yourself, rather than only adding features to an existing stack
• Distributed systems, reliability and fault tolerance are second nature; you're excellent at database design and performance tuning
• Strong architectural opinions, and the backbone to rebuild what's broken (and defend why)
• You take ownership when things are fast and chaotic
• AI-native in how you work
• Genuine curiosity about mortgage and lending
• Based in SF or ready to relocate right away; 5 days a week in person

Where you might come from

• Distributed systems or low-latency backend at high-bar infra and data companies (databases, streaming, search, vector stores, cloud compute layers)
• AI or voice infrastructure: production infra at AI labs and inference platforms, voice AI and telephony infra, sub-second systems
• Founder or founding engineer at a startup with serious backend architecture
• Infra teams at fast-moving AI startups, regulated vertical AI or lending tech
• Big-tech infra, with clear zero-to-one signal

Bonus points

• Kubernetes and event-driven, parallel architectures in production
• Low-latency systems where every millisecond counts: voice AI, high-frequency trading, ad tech, real-time bidding