Agentic DevOps Engineer

About Atlas

Atlas is a CRMx for the agency recruitment industry. We give recruiters a living memory of everything they say, hear, read, and write — and we're equipping them with full digital workers to make them tomorrow's market leaders.

We're backed by Smedvig Capital and founded by repeat entrepreneurs whose previous company is doing $50M in revenue today.

What makes this opportunity extra special is that around 40% of the company is pledged to the team. Not options but the underlying ownership in the company.

We believe deeply in our values and expect anyone joining us to be aligned with them: https://bit.ly/recruitwithatlas-values

The Role

This isn't a traditional DevOps role. Like our platform, every team member is agentic AI first and so we expect this hire to continue our work building agentic systems that monitor, diagnose, and heal themselves, If you've looked at an alert, thought this should just fix itself, and built an agentic system to do so, this is the place for you.

You'll own our infrastructure as we scale from thousands to hundreds of thousands of users. We're already on a solid foundation, Elastic for search, read replicas, a horizontally-ready architecture. You're job is to ensure our platform scales with growth.

Our team is made up of only senior engineers and so no mentoring is expected. Instead, we prefer the diffusion of knowledge amongst skiller practioners.

What You'll Be Doing

• Design and build self-monitoring systems that detect anomalies, correlate signals, and surface root causes without a human in the loop
• Build agentic remediation pipelines that resolve common scaling failure modes, query degradation, replica lag, memory pressure.
• Own horizontal scaling as we grow, with a clear runway to work from
• Work closely with a product engineering team of 12-13 to ensure infrastructure never becomes the constraint on delivery speed

Our stack

• Our applicants are typescript (node/react) with Python microservices
• We use AWS, Postgres and Elastic.
• We are also using Redis, BullMQ and Debezium.
• For logging we use NewRelic and Rollbar

Who You Are

• You've scaled a SaaS product through the 1,000–10,000+ user range and have the playbook
• You think in systems, not tickets, you'd rather build the thing that prevents the problem than fix the problem repeatedly
• You have genuine experience with agentic or AI-powered infrastructure tooling, or a strong point of view on how to build it. You've probably already spent your allocated AI budget!
• You're comfortable working in a fast-moving environment where priorities are real and the team is small
• You are a scientist — you'd rather figure things out yourself than wait to be to
• You dislike meetings and prefer the mutual trust of an experienced team that have done it before
• You are the engineer that no-one wants to lose.
• You've stuck around in your prior companies and have a clear uwards trajectory (start-ups are hard and we want someone who has stuck it out!)

In order to be considered, please make sure you have clear evidence for the above in your resume.

Compensation

• We pay above market and will structure what's necessary.
• c40% of the company equity has been given to the team. This will be the largest equity cheque you'll ever receive outside of founding your own company

Location

We will hire a superstar wherever they are, but prefer to add engineering heads to countries where we already have a presence

• India
• Bosnia
• Ukraine
• UK
• Italy

Interview Process

1. Interview with a Founder (45 mins)
2. Mini IQ test (15 minutes)
3. Interview with VP Engineering (60 minutes)
4. Technical case study (90 mins)
5. Offer

Note on AI applications: We are scanning inbounds for obvious AI. While we don't mind people using AI for submissions, forcing keywords into a resume without the underlying requirements is not what we are looking for.