Staff Software Engineer, Platform

About Enterpret

At Enterpret, we are building the customer feedback intelligence platform that turns unstructured customer feedback into a structured, queryable source of truth — solving complex problems in natural language processing, serverless computing, and analytics on the frontend, pushing the envelope of what's possible by applying first principle thinking.

Kleiner Perkins and Sequoia Capital back us because they share our conviction that product development teams deserve a better way to understand their customers. That conviction is validated by the respected product teams who rely on us every day, including Canva, Notion, Samsung, and Loom.

We love working with folks who are resourceful, thrive in ambiguity, and display a strong sense of ownership.

Our engineering culture is built around leveraging and contributing to open-source tools. We aim to build upon and improve state-of-the-art systems in our field of work.

Read more about our team, core values, and operating principles - here .

What You'll Do

Be the owner for Graph Platform — own the knowledge graph, the data systems behind it, the Model Repository, and the inferencing and enrichment pipelines end to end, from design through production operation.

Design for where our scale is headed in the next 1–2 years and work backwards — most decisions in this part of the stack are hard to reverse, so the judgment to make them well is the core of the job.

Upgrade the indexing layer (DynamoDB, ClickHouse, Snowflake) and the inferencing layer (model hosting and pipelines) so enterprise and PLG clients get the same experience on latency and availability, regardless of data volume or query throughput.

Take production issues beyond the symptom — find the underlying disease and cure it for the long term , then build the metrics, alerts, runbooks, and rollback thinking that keep it fixed.

Make sound calls on which systems need an autoscaling-policy update and which need a re-design — and know the difference.

Own cost vigilance on the company's most expensive cloud resources: establish a baseline for cost per feedback record and drive it down.

Partner with product engineers and PMs on upcoming use cases, with SRE on AWS infrastructure, and mentor engineers as the team grows.

What It Takes

8+ years in software engineering, with deep experience designing and operating data-intensive systems in production at scale.

3+ years at one organization where you've owned production databases and infrastructure through real growth — not just built them, but scaled and operated them under load.

Extensive hands-on experience running DynamoDB and ClickHouse at scale , strong command of data-intensive application design , and comfort working close to AWS infrastructure .

You think in the terms of Designing Data-Intensive Applications and The DevOps Handbook — first-principles design, engineering metrics, and operational rigor are how you already work.

Proven judgment on 1–2 year, hard-to-reverse decisions : you know when an autoscaling-policy change is enough and when a system needs a re-design.

Strong ownership and the energy to operate in chaos, plus an AI-native working style — you move at a much higher pace without compromising platform integrity or quality gates.

A track record of mentoring and collaborating — you're the engineer a growing team can grow around. Prior self-hosted-LLM experience is not required; the curiosity and speed to learn and run new areas at scale is.

What Success Looks Like

First 90 days

Gets hands-on across the stack and learns how the systems behave under real load.

Takes a production issue end to end — finds the disease beyond the symptom and cures it for the long term.

Makes sound calls on which systems need an autoscaling-policy update and which need a re-design.

By 6 months

Is operating as the Graph Platform owner in practice and owns the work end to end.

Has an executable plan underway for the indexing …