Engineering Manager

CloudBees; Chennai District; On-site; INR 3,000,000 - 6,000,000; Full time

Job summary

CloudBees is seeking a senior engineering leader to own delivery and team health for Feature Management across a distributed team. You will drive architecture decisions with your engineers and foster rapid feedback loops, not waiting for annual reviews.

You’ll hire, coach, and have tough conversations to keep performance aligned with goals. You’ll lead a cross-continental group, clear bottlenecks, and push for AI-assisted workflows to accelerate delivery while maintaining quality and governance.

Qualifications

• Track record of improving automation and developer tooling.
• 5+ years as software developer before moving into management.
• 5+ years as engineering manager with 5+ direct reports.
• Hands-on CI/CD, feature flagging, or platform engineering experience.
• Experience leading distributed, multi-timezone teams.
• Appetite for AI in coding and using coding agents.

Responsibilities

• Own delivery and team health for Feature Management.
• Steer decisions with questions, then empower engineers to own the architecture.
• Provide feedback in the moment, not in a post-mortem essay.
• Hire, grow, and have hard conversations when performance is off.
• Cut unnecessary processes that exist due to fear.
• Use AI agents to run your own work and track team metrics across tools.

Skills

Automation

Developer tooling

Engineering management

CI/CD

Feature flagging

Platform engineering

Distributed teams

AI in coding

People management

Decision making

Tools

Go

Job description

CloudBees helps organizations build, run, and govern software factories, giving enterprises the confidence to ship software better, faster, and safer.

Writing code is no longer the bottleneck. Governing what reaches production and validating its impact is. As enterprises adopt agentic coding, software is created faster than most teams can review, secure, and validate. Without consistent governance, organizations risk shipping code they cannot explain, audit, or trust.

CloudBees addresses this challenge without asking teams to replace the tools they already use. Across every toolchain a customer runs, CloudBees makes each change, human or AI, visible, auditable, and accountable before it reaches production. CloudBees Unify is the product behind this: a governance layer, with context and control-plane capabilities, that enforces consistent policy and evidence across every tool, team, and workflow.

Founded in 2010, CloudBees is backed by Goldman Sachs, Morgan Stanley, Bridgepoint Capital, HSBC, Golub Capital, Delta-v Capital, Matrix Partners, and Lightspeed Venture Partners.

CloudBees builds CloudBees Unify, a governed software delivery platform spanning CI/CD, release orchestration, and feature management. Feature Management is the capability that lets teams ship code with a feature turned off, roll it out gradually, and kill it instantly if something breaks, no redeploy needed.

This team is moving into a new org, and that's where you come in.

You'd inherit a team spread across three continents and about ten time zones, and likely pick up one or two additional small teams alongside it. The team includes engineers who do not need you to tell them how to write code. What they need is someone who clears the path between them and shipping and treats "I don't know, let me find out" as a perfectly good answer in a 1:1.

This is the role today. If you do it well, there's room to take on more.

What you'll do

• Own delivery and team health for Feature Management
• Steer decisions with questions, then let your engineers own the architecture. If you're the one drawing the boxes, something's gone wrong
• Give feedback in the moment, not in a surprise essay once a year
• Hire, grow, and have the hard conversations when someone isn't where they need to be
• Cut process that exists because someone was once scared
• Use AI agents to run your own job faster: pull together who's on PTO next week, the lead time of your team's PRs, what ticket needs attention, what shipped and what it broke, instead of digging through five tools by hand

Not write code. Not approve every PR. Not sit in on every incident call because you don't trust the on-call rotation.

Requirements

What we're looking for

Must have

• A track record of improving automation and developer tooling, not just using what's handed to you
• At least 5 years as a software developer before moving into management, so the technical instincts below are earned, not borrowed (ideally, you have developed in Go)
• 5+ years as an actual engineering manager, with 5 or more direct reports at once. Not a tech lead with 3 reports and half a week of coding
• Genuinely technical: enough hands‑on background in CI/CD, feature flagging, or platform engineering to ask a sharp question in a design review, plus real experience managing engineers who are better coders than you, and you're comfortable with that
• A track record of running distributed, multi-timezone teams: clear owners, clear deadlines, decisions that don't wait on a meeting that nine time zones won't allow
• You take the heat for your team and hand them the credit
• The instinct to measure what shipped and what it did for customers, not green dots in Slack. You know deployment frequency, lead time, change failure rate, and time to restore, and you use them to start conversations, not to rank people
• A real appetite for AI: recent, hands‑on use of at least one coding agent (Claude Code, Gemini, Cursor, Kiro, or similar) to run your own work, not necessarily to ship product code. You push your team to try these tools too, and have opinions about where they help and where they don't, not secondhand takes from a keynote you watched once
• Experience taking over a team mid-transition

Nice to have

• A sense of humor. You'll need one some days, and it'll help on the good ones too
• Prior exposure to release orchestration

Who this is for

You'd rather ask what's actually blocking someone than "why isn't this done yet." And when the unpopular call is needed, you make it, even when the timing is inconvenient.

You're fine with ambiguity that has a shape, not ambiguity that's just chaos with a Jira board. Sharp and a little irreverent beats technically flawless but rigid, as long as you still ship.