Founding Platform Engineer - AI Startup

Founding Platform Engineer

SUMMARY

Company Stage

Pre-Seed 1M

Employer Tech Stack

Python, AWS, Postgres, MongoDB

Acceptable Tech Background

Distributed

Locations

This role will be fully remote for ~1 year then move to hybrid in either location.

San Francisco (San Francisco Bay Area)

New York City (New York Area)

QUICK PITCH

Provides AI monitoring for customer conversations across voice, chat, email, and SMS.

The Role

You'll be the first senior engineer and the architectural counterpart to the CTO. You'll write code regularly, own the hardest parts of the platform, and set the bar for the engineers who build alongside you.

What you'll own

• Conversation pipeline and integrations: Normalization, segmentation, and enrichment that turn raw conversations into structured, scorable evidence, including the audio and transcription path. You'll build the first contact-center, ticketing, and direct-upload integrations, plus the reusable framework a future integrations engineer will extend.
• Provenance and auditability: Immutable versioning for rubrics and scoring configuration, tamper-evident audit trails, and the ability to answer what an interaction scored eight months ago and why.
• Multi-tenant platform: Tenant isolation from the infrastructure level down, per-customer configuration and entitlements, and the roles and permissions enterprise data separation requires.
• Engineering platform: IaC, CI/CD, environment management, observability, secrets, backups, disaster recovery, and incident response. There's no dedicated DevOps or SRE, so every engineer owns the production behavior of what they ship.
• Data and reliability: Storage and retention at volume, queueing, idempotent reprocessing that never multiplies inference spend, on-call, SLOs, and a release process that doesn't depend on heroics. Getting from pilot scale to millions of interactions a month is part of the job.
• Security and compliance: Sensitive-data handling and redaction before any model sees a conversation, access control, evidence export, and the engineering half of SOC 2 and enterprise security reviews.

What we're looking for

• Production ownership at staff or principal scope, including systems other teams depended on and the operational burden that came with them
• Strong distributed-systems and backend depth, with high-volume data or event-processing experience and solid API and data-model design
• Experience owning infrastructure without a DevOps team, having built the IaC, CI/CD, and observability layer yourself
• Startup experience at an early-stage or high-growth company (required, not a preference)
• Depth in Python, plus enough TypeScript and React to make small frontend changes yourself
• Fluency across the ML pipeline, enough to discuss accuracy, latency, and cost with the ML Lead (no research work involved)
• Multi-tenant SaaS experience at enterprise scale, including tenant isolation and per-customer entitlements
• Real leverage from coding agents (the team uses Claude Code daily) and the judgment to hold a quality line on their output

Bonus

• Audit, provenance, or compliance-grade data systems
• Speech and audio in production (ASR, diarization, streaming transcription, real-time latency budgets)
• Contact center, CCaaS, QA and workforce optimization, or compliance and GRC domain experience