Member of Technical Staff, Applied AI Backend

Mercor•San Francisco, CA

21h•$220,000 - $500,000•Onsite

About The Position

The Applied AI org builds the systems that turn human expertise into training data for frontier models, task pipelines, expert workflows, evaluation infrastructure, and the services that tie them together. We have synthetic pipelines and modular quality control systems that run in unison to generate highest quality tasks and at scale. All of it runs on backend systems that have to stay correct, fast, and observable while the volume behind them grows every month. As a Software Engineer on Backend Systems, you'll own services in that stack: designing the data models, building the APIs, services, data pipelines that move work through the platform. This is a build role. You'll take a problem that's roughly scoped, make the design calls, ship it to production, and own it afterward. We're hiring for depth in backend fundamentals rather than any particular domain. If you've built and operated real services, handled the schema migration that couldn't take downtime, found the query that fell over at 10x traffic, designed the retry logic that made a flaky dependency invisible to users, or build a system that recover when things break that's the experience that matters here. You'll work with a talent dense group of engineers who will review your designs and push your thinking, and you'll be expected to grow into owning larger surfaces quickly.

Requirements

2–5 years of professional backend engineering experience building and operating production systems.
Strong fundamentals in backend engineering: data structures, algorithms, concurrency, and writing code that's clear enough for the next person to change.
Hands-on experience with API design - REST, gRPC, or GraphQL, and an understanding of versioning, contracts, and backward compatibility.
Solid database skills: relational data modeling, indexing, query performance, transactions and isolation, and safe migrations. Familiarity with at least one NoSQL or key-value store and when it's the right choice.
Practical experience with distributed systems basics: queues and event streams, caching, idempotency, rate limiting, and designing for partial failure.
Experience with Data Orchestration and workflow management systems like Airflow, Temporal, Dagster.
Be able to roll your sleeves up and dig deeper into the lower level infrastructure issues container, permissions, logs, traces and find the needle in the haystack.
Experience running services in production: containers, CI/CD, monitoring and alerting, and debugging issues under real traffic.
Comfort with ambiguity you can take a loosely defined problem, ask the right questions, and come back with a plan.
Genuine excitement for agentic development and new technology, fluency with modern AI dev tools (e.g. Claude Code, Cursor, Copilot), and a real passion for writing good code.
Clear written and verbal communication. High ownership, pragmatism, and a bias toward shipping.

Nice To Haves

Experience building or integrating with LLM-backed services in production (evaluation, orchestration, or serving).
Familiarity with AI infrastructure providers like Modal, Fireworks, Baseten, Temporal

Responsibilities

Design, build, and operate backend services in the Applied AI stack including APIs, data models, background jobs, and the pipelines that connect them.
Own features end to end: scope the problem, write the design, ship the code, instrument it, and keep it healthy in production.
Build and tune high-throughput data and job pipelines: queuing, batching, idempotency, retries, and backpressure.
Make the system fast and reliable by adding failure recovery in pipelines, profile hot spots, Agent token and cost attribution, caching issues, and set latency, error, cost budgets you actually hold to.
Instrument what you ship: observability, metrics, logging, tracing, and alerts that proactively catch problems.
Work on modern cutting edge tools, libraries and frameworks.
Manage and launch 10s of 1000s of containers, sandbox environments, manage resource allocation and system health.
Manage infrastructure as code using Terraform.
Participate in on-call for the systems you own, debug production incidents, and write up what you learn in RCCA.
Write clear design docs and give useful code review, we make technical decisions in writing and expect everyone to take part.
Work directly with product, operations, and research partners to turn ambiguous requirements into systems that ship.