Senior Backend Engineer - Developer Experience & Productivity

Einblick in das Angebot

About the Role
Taktile helps financial institutions make smarter, safer decisions with the power of AI. Internally, AI coding agents have made writing code dramatically faster. The bottleneck is now everything around the code: builds, tests, reviews, merges, and deploys.
We're looking for a Senior Backend Engineer focused on Developer Productivity and Developer Experience to remove that bottleneck. This is a high-leverage, internal engineering role on our Platform Team: your users are Taktile's backend and frontend engineers and the AI agents working alongside them, and your product is the path from first commit to production.
You'll make CI fast and reliable, make our serverless stack easier to build on, and turn "how fast can we ship safely?" into metrics we move every quarter.

What you’ll do
Build the deployment services that ship Taktile. Own the systems behind tens of thousands of stack deployments a week across dev, staging, production, and per-customer AWS accounts: progressive rollouts through monitoring and canary stages, deployment windows and gates, and fast rollbacks.

Keep our AI developer tooling sharp. Maintain and extend Claudius, our in-house AI engineering agent that investigates CI failures, triages issues, and opens PRs, and metadev, our sandboxed Claude Code and OpenCode setup that packages Taktile's engineering knowledge for every engineer.

Own the build → test → merge loop. Drive down CI runtime and flakiness across our main monorepos. Find the slowest and least reliable steps, fix them, and keep them fixed.

Evolve our CI/CD architecture. Build on our move to per-module GitHub Actions workflows and a self-hosted runner fleet. Improve caching, parallelism, and cost per build.

Shorten the inner dev loop. Make developing on our AWS Lambda stack fast, from running services locally against preview environments to faster, container-based deploys, so engineers aren't waiting on CloudFormation to test a change.

Build paved roads. Ship templates, CLIs, and self-service tooling so any team can take a new module from first commit to a running dev workspace in under a day.

Design for humans and agents. Make our tooling fast, deterministic, and scriptable so engineers and AI coding agents get quick, trustworthy feedback.

Measure what matters. Own DevProd metrics like PR lead time, CI runtime, CI flakiness, and PRs per engineer-day. Use them