Member of Technical Staff, Core Backend

Vapi; San Francisco (CA); On-site; USD 120,000 - 160,000; Full time

Benefits offered by this job

Comprehensive health coverage

Flexible time off

Catered meals

Transportation

Quarterly off-sites

Annual L&D budget of $10k

Job summary

Vapi, based in San Francisco, is hiring for a role focused on the StreamModule pipeline that powers voice AI. You will be responsible for ensuring the pipeline's reliability and enhancing its capabilities without affecting the core architecture.

The ideal candidate should have extensive experience with real-time systems and queue architectures like BullMQ and Kafka, along with a solid understanding of Postgres management and OpenTelemetry instrumentation.

Join our dynamic team where 70% of employees are former founders and enjoy competitive salary, equity, and an excellent benefits package.

Qualifications

• Experience with real-time or streaming systems in production.
• Knowledge of queue architecture and when to use each type.
• Strong background in building plugin or adapter architectures.
• Experience managing Postgres at scale, including read replicas.
• Ability to instrument systems using OpenTelemetry.

Responsibilities

• Ensure pipeline stability and pluggability for voice agents.
• Consolidate BullMQ into Kafka for improved messaging.
• Implement event-driven instrumentation for pipeline stages.
• Drive improvements in reliability and latency on the call path.
• Collaborate with design reviews on pipeline changes.

Skills

Real-time or streaming systems

Queue architecture experience (BullMQ, Kafka)

Plugin or adapter architectures

Postgres operating experience

OpenTelemetry instrumentation

Tools

TypeScript

Node.js

NestJS

Liquibase

Job description

Overview

Voice AI that resolves, not transfers. Most phone systems trap callers in menus and scripts. Vapi is the platform for deploying voice agents that know your business and can listen, adapt, and resolve in minutes.

Why We’re Hiring This Role

The StreamModule pipeline — VAD → STT → LLM → TTS → Transport — runs on cork/uncork backpressure during live phone calls. A hundred milliseconds of delay is audible. This role owns pipeline stability and pluggability, so the agents and FDE teams can add new models and providers without touching core.

You’ll consolidate BullMQ into Kafka, harden the provider abstractions (LLM, STT, TTS base classes), instrument the pipeline with event-driven OTEL tracing, and shore up the Postgres SPOFs that contributed to the Oct 15 and Oct 22 incidents.

What You’ll Do

30 Day: Ramp on the StreamModule pipeline and the cork/uncork backpressure model. Walk the Oct 15 / Oct 22 DB incidents and the duplicate-message incident. Land a scoped pipeline or provider-abstraction improvement.

60 Day: Own a slice of the BullMQ → Kafka consolidation. Ship event-driven OTEL instrumentation for at least one critical pipeline stage. Harden one provider plugin path so a new model can be added without core changes.

90 Day: Drive a measurable reliability or latency win on the call path. Be the backend owner that agents and FDE teams pull in for design reviews on new providers and pipeline changes.

Must-haves

Who You Are:

• You’ve built real-time or streaming systems in production — media pipelines, streaming data, or event-driven backends. You’ve debugged a backpressure cascade.
• You have opinions on queue architecture (BullMQ, Kafka, Temporal) and when each is the right fit.
• You’ve built plugin or adapter architectures — extending base classes cleanly, with decoupled implementations.
• You’ve operated Postgres at scale: connection pooling, read replicas, schema migrations (Liquibase or similar).
• You instrument with OpenTelemetry and think in event-driven traces, not just logs.

Nice-to-haves

• TypeScript + Node.js + NestJS. The codebase is huge NestJS, but a strong systems-thinking engineer ramps fast — language doesn’t gate the hire.

Tech Stack You’ll Work In

• Languages: TypeScript on Node.js (primary).
• Framework: NestJS (large codebase).
• Pipeline: StreamModule (VAD → STT → LLM → TTS → Transport), cork/uncork backpressure.
• Queues: BullMQ (current), Kafka (target — consolidation on roadmap), Temporal.
• Database: Postgres (connection pooling, read replicas), Liquibase for schema migrations.
• Plugin system: provider abstractions — LLM, STT, TTS base classes (pluggable, decoupled from model integrations).
• Observability: OpenTelemetry tracing, event-driven instrumentation.

Where You Likely Come From

A streaming or real-time platform (Discord, Slack, Zoom, Twitch, Mux, LiveKit), an ML-infra company (Modal, Baseten, Replicate, Together), or a pipeline/workflow shop (Temporal, Stripe Radar, trading systems).

Weak fit: backend engineer who’s only built systems where users don’t wait in real time (overnight jobs, reports, dashboards).

Why Vapi

Generational impact: Build the human interface for every business

Ownership culture: 70% of the company are previous founders

Kind team: The founders, Jordan and Nikhil, are Canadians

Tier-1 Investors: YC, KP seed, Bessemer Series A, Recent Series B raise

What We Offer

Real stake: We offer a competitive salary and excellent equity ownership

Comprehensive health coverage: medical, dental, and vision plans

Team love: We love hanging out, and we do quarterly off-sites

Flexible time off: take what you need

More: catered meals, transportation, gym, and a $10k annual L&D budget