Skip to content
Search jobs inside your AI agent. Claude, ChatGPT, Claude Code, Codex, Gemini, Cursor and Grok
Connect

JoBins

NewBe an early applicant

DevOps / Platform / SRE Engineer

Posted Updated

Summary

Hands-on DevOps/SRE engineer owning AWS production infrastructure (EC2, EKS, RDS MySQL, ElastiCache, OpenSearch, Cloudflare) for a Japan-based recruitment platform. Day to day: monitoring and incident response, fast CVE patching, Terraform IaC, CI/CD, developer self-service, and AWS cost control, with fair on-call.

JoBins (jobins.jp) is a recruitment platform in Japan that connects hiring companies with recruitment agencies. Alongside our core recruitment platform, we are expanding our AI-powered product ecosystem, including AI Agent Mode, AI Screening, Job Extraction, and Resume Parsing.

Thousands of companies and recruitment agents rely on our platform every business day.

As our products and infrastructure continue to grow, we are looking for a hands-on DevOps / Platform / SRE Engineer to take ownership of our cloud infrastructure, strengthen production reliability, and build a secure and scalable platform for our engineering team.

This is an opportunity to work on real production challenges, take technical ownership, and directly influence the reliability, security, and performance of a growing technology platform.


About the Role

You keep production up, secure and affordable, and you turn our "it works on that one EC2 box" setup into a platform the developers can ship on safely.


This is a hands-on role. One day you're rolling out a PHP patch across every host before lunch. The next you're finding out why 50 FPM workers are all waiting on one MySQL query, and then writing the Terraform so the same mistake can't happen twice. Uptime is the job. Making the next incident boring is the career.


If you think chmod 777 is a valid fix, latest is a version number, or "we'll add monitoring later" is a plan, this probably isn't the role for you. If those sentences made your eye twitch, keep reading.


What you'll do

Run production (SRE)

  • Own the reliability of jobins.jp, the API and the AI services: EC2, EKS, RDS MySQL, ElastiCache, OpenSearch and Cloudflare.
  • Set up and tune monitoring, logging and alerting so we hear about problems from a dashboard, not from a customer.
  • Lead incident response: find the root cause, write a blameless postmortem, and make sure the fix actually ships.
  • Join a fair on-call rotation, especially around the 9–11 JST peak.


Security, at godspeed

  • Track CVEs and vendor advisories across our stack (PHP, Laravel, Node/Next.js, Kubernetes, base images, OS packages).
  • When a critical one drops, patch every affected host and container within hours, not weeks. Recent example: a pre-auth PHP RCE with public exploits on day one.
  • Harden things before anyone asks: least-privilege IAM, no static keys where roles will do, private networking for data stores, origin lock-down behind the WAF, MFA everywhere, secrets in a secrets manager rather than in Slack.
  • Own credential rotation, access reviews, and onboarding and offboarding for infrastructure access.


Build the platform

  • Move us from hand-configured servers to infrastructure as code (Terraform or similar) with reviewed changes.
  • Improve CI/CD: safer zero-downtime deploys, image builds with scanning, and fast, boring rollbacks.
  • Grow our Kubernetes (EKS) footprint where it makes sense, and say clearly where it doesn't.
  • Give developers self-service paths: staging environments, preview apps, sane defaults, and docs they will actually read.
  • Work with developers on performance: slow SQL, missing indexes, cache stampedes, queue workers, FPM tuning.


Keep the bill honest

  • Know what every major line on the AWS bill is for, and right-size, reserve or delete accordingly.
  • Weigh every new tool against what it costs to run and maintain. Sometimes the right answer is a cron job, not a new cluster.


Must-haves

  • 3+ years running production systems in DevOps, SRE, platform or infrastructure work. (Strong 2-year people who've already been paged at 3 a.m. are welcome to apply.)
  • Solid AWS experience: VPC, security groups, IAM, EC2, RDS, ElastiCache, ALB, CloudWatch. You can explain why the database shouldn't have a public endpoint.
  • Comfortable on Linux (Ubuntu): systemd, Nginx, logs, ss, top, journalctl, and the patience to read them.
  • Infrastructure as code (Terraform, OpenTofu, CDK or Pulumi) and CI/CD pipelines you've built and maintained.
  • Containers and Kubernetes in production, not only in tutorials.
  • Security instincts: patch management, least privilege, secrets management, and a habit of asking "who else can reach this?"
  • Working knowledge of MySQL/Postgres: reading slow query logs and EXPLAIN, and knowing what an index is for is a plus point.
  • Fast response and good availability, especially when a critical CVE or an outage lands. Calm under pressure, clear in a Slack incident channel.
  • Cost sense: you can say what a change costs per month and whether it's worth it.
  • Good written English. Clear runbooks and postmortems are part of the job.
  • Hands-on use of AI coding agents in real infrastructure work, with clear rules for keeping them away from secrets and production write access (see the next section).


Nice-to-haves

  • PHP/Laravel operations: PHP-FPM tuning, OPcache, Horizon, Laravel Forge.
  • Cloudflare: WAF rules, Zero Trust / Access, tunnels.
  • Redis/Valkey and OpenSearch/Elasticsearch operations, including k-NN and vector workloads.
  • Running AI/LLM services in production: GPU or inference cost control, API key governance, usage budgets.
  • Observability stacks: Prometheus, Grafana, OpenTelemetry, Datadog or similar.
  • Security tooling: GuardDuty, Security Hub, trivy, gitleaks; experience with SOC 2 or ISMS (ISO 27001).
  • Kafka or other event streaming.
  • Certifications such as AWS Solutions Architect, CKA or CKS are a plus, not a requirement. We care more about what you've fixed than what you've passed.


You'll thrive here if you…

  • Are proactive. You read the security advisory before it trends on Hacker News, and you open the ticket before the incident.
  • Learn fast. New tool, new cloud service, unfamiliar codebase: you get productive in days, not quarters.
  • Love new tech and also know when to say no. You're excited to try things, but you can explain the trade-offs, the cost and the on-call burden of each choice. "Boring and reliable" is a compliment.
  • Think in root causes. Restarting the service is a stopgap, not a fix. You want to know why.
  • Automate the second time. Do it by hand once, write the runbook the second time, automate it the third.
  • Communicate clearly. Developers, tech-leads and non-engineers all understand what you changed and why.
  • Practice blameless ownership. Everyone breaks prod eventually. What matters is the postmortem and the fix.

You probably won't enjoy it here if you want a strict ticket queue and nothing else, you'd rather rewrite everything than improve what exists, or you think security is someone else's department.


What we offer

  • Real ownership. Your decisions show up in production, on the bill and in fewer 3 a.m. pages.
  • A small team that makes decisions quickly. No change advisory board sitting on your pull request for three weeks.
  • Real problems at a real scale: high-traffic Laravel, AI and vector search workloads, and multi-account AWS security.
  • Fair on-call: a rotation, compensation for it, and a team that fixes the root cause so the same page doesn't wake you twice.


Role Type: Full-time (Remote)

Location: Nepal

Working Hours: 7:00 AM – 4:00 PM NST (Monday to Friday).


Additional Benefits:

  • Quarterly team-building events.
  • Quarterly performance-based bonuses.
  • Semi-annual performance reviews.


Interested?

If you enjoy solving complex infrastructure challenges, building reliable systems, and taking ownership of production environments, we'd love to hear from you.

Skills

Apply

See also

DevOps jobs by country — openings, pay and top skills →