Senior Software Engineer, Developer

Sprig is the customer survey platform, rebuilt around AI agents. For two decades, teams have relied on Qualtrics, Medallia, and Survey Monkey to understand their customers. These platforms were built for a slower era, with manual setup, fragmented workflows, and insights that arrive too late to share decisions. The gap between how fast products move and how fast research can keep up has never been wider.

We’re building something different. With Sprig, AI agents design studies, adapt surveys in real time, and turn responses into clear, actionable insight. Teams move from question to evidence in hours instead of weeks, without sacrificing rigor. Product teams no longer guess. They know. Our mission is to make deep customer understanding effortless and always on, so the best teams in the world can build products their customers love.

Companies like Microsoft, Door Dash, Notion, Figma, Coinbase, and Trip Advisor use Sprig to stay close to their users. Backed by $85M from Andreessen Horowitz, Accel, and First Round Capital, we’re growing quickly across the world’s leading product organizations. The category is being rebuilt, with AI at the core. If you’re excited by hard problems and the chance to define how modern teams understand their customers, we’d love to meet you.

Agents pay off when engineers hand them whole changes, not incremental changes. That makes the build-and-test loop the safety gate for the entire org: when CI is slow or flaky, teams lose time twice, once waiting and once re‑running to sort noise from real failures. Add a dozen agents validating against the same shared environment, and a workflow annoyance becomes an organization‑wide bottleneck.

This role owns that loop. You will be responsible for how quickly a change becomes a signal people trust, whether a human or an agent wrote it: incremental CI on a large Type Script, React, and Go monorepo, test automation that is fast and meaningful from unit through browser and visual, and short‑lived environments created and torn down per branch instead of queued.

We are not looking for someone to inherit a ticket queue. We want a builder who can help set a direction for how development should work in an agent‑heavy world, then deliver it. If you have experience making a development loop people complain about faster and more predictable, this is the next iteration of that same problem, with tighter constraints.

You will do this on the Platform team, which runs the systems that support over a billion visitors a month and processes nearly a billion events a day. This is a hybrid role based in our San Francisco office 3x per week (M, T, Th).

• Modernize the JavaScript and Type Script build:
  Lead the migration from our current Node Bazel rules to a maintained successor, complete the move of the JS toolchain to Bazel modules, and replace accumulated image and packaging workarounds with proper module linking and incremental compilation.
• Cut CI time and cost, measurably:
  Establish p50 and p95 PR build and test times as tracked metrics, then drive them down through better cache hit rates, incremental builds, and balanced test parallelism. Own the remote cache fleet's sizing, eviction behavior, and cost.
• Make the build observable:
  Ship telemetry for per-target timing, cache hit rate, flake rate, and cost, so build decisions are data-driven rather than anecdotal.
• Make test automation a first-class developer experience:
  Own how engineers and agents write, run, and trust tests across unit, integration, browser end-to-end, and visual layers. Make the local command match CI, make it trivial to run either a single file or the affected graph, and detect and quarantine flakes instead of asking people to rerun.…