Remote, India · Full-time · Typically 3+ years of relevant experience

Reporting to: CTO · Infrastructure: self-managed production servers hosted in the United States, no managed public-cloud services

This is a hands-on infrastructure role, not a research or model-training role. "AI" in the title means two things: you run the infrastructure behind Natoe's AI-assisted radiology products, and you use AI tools daily to automate operations while owning the security of everything they produce.

About Natoe AI

Natoe AI is an AI-native teleradiology company combining radiologist reporting with technology for medical-imaging and reporting workflows. Imaging centres send us studies, radiologists read them, and reports go back, often on urgent turnaround. Our infrastructure handles sensitive U.S. healthcare information and large volumes of DICOM imaging data.

Why this role exists

AI now writes a large share of production code, including ours, and that has made infrastructure ownership more important, not less. Independent 2026 studies find that close to half of AI-generated code samples contain security flaws, and software vulnerabilities are now the leading breach entry point. Tools can generate a script in seconds; someone still has to decide whether it is secure to run against patient data.

You will be that person for Natoe: the accountable owner who keeps imaging and reporting services secure, reliable and recoverable, and who makes our engineering team's use of AI safe by design.

Responsibilities

- Secure delivery of AI-written code. Build the guardrails every change passes before production: static analysis, secrets scanning, dependency and container-image checks, and review rules for high-risk paths.

Track how often AI-assisted changes fail these gates and tighten them over time.
- Governance of AI tools and agents. Define what AI assistants and automation agents may access, with scoped, short-lived credentials and audit logs. Keep patient data, credentials and sensitive logs out of unapproved tools.
- Infrastructure for Natoe's AI products. Run and scale the servers, storage and pipelines that serve our AI features alongside radiologist workflows, including capacity planning as model usage grows.
- AI-assisted operations.



Use Claude Code, GitHub Copilot or equivalent to automate provisioning, patching,

deployment checks, monitoring, backup verification and audit-evidence collection. Review, test and maintain everything generated, under normal change control.
- Dedicated-server operations. Provision, harden, patch, upgrade and troubleshoot Linux servers. Operate database and container infrastructure with engineers. Coordinate hardware incidents and remote hands with hosting providers.
- Networking and secure connectivity. Manage firewalls, VPNs (including IPsec links to imaging centres),

DNS, certificates, reverse proxies, load balancing and controlled admin access. Keep production, staging and development separated.
- Medical-imaging storage and data integrity. Operate DICOM and object storage: capacity, throughput,

retrieval, caching and archiving. Detect failed transfers, corrupted objects and file-to-database mismatches.

Implement approved retention, deletion and data-location rules.

- Reliability and incident response. Monitor image transfer, processing backlogs, viewer availability and report delivery. Lead infrastructure incidents, communicate impact, and document root causes and fixes.
- Security controls and audit support. Implement access controls, MFA, encryption, secrets management,

security logging and vulnerability remediation. Maintain infrastructure evidence for HIPAA, SOC 2 and customer reviews. Policy, contracts and risk acceptance stay with the CTO and compliance owners.
- Backups and disaster recovery. Keep protected off-host backups separated from production credentials.

Agree RTO/RPO with the CTO, run timed restores with database-to-image consistency checks, and test recovery from server failure, deletion, ransomware and provider outages.
- Safe deployments.



Build repeatable configurations and pipelines with version control and Ansible or equivalent, with pre-deployment checks, rollback criteria and post-change validation.
- Continuity and leadership. Maintain architecture diagrams, inventories, access records and vendor contacts. Keep runbooks usable by another engineer, mentor colleagues and set up planned backup coverage.

Required experience

- Typically 3+ years in infrastructure engineering, systems administration, DevOps or SRE, with independent ownership of dedicated-server, bare-metal or comparable self-managed production environments.

Equivalent hands-on depth will be considered.
- Strong Linux, networking, firewall, filesystem and storage troubleshooting, plus production support for databases and containerised applications.
- Substantial file or object-storage operations: capacity, performance, integrity and migrations. You have personally restored from backup or led a disaster-recovery exercise.
- Bash or Python, version control, CI/CD and configuration automation with Ansible or equivalent.
- Daily, hands-on use of AI coding assistants, with the judgement to explain, debug and reject generated code.
- Hands-on DevSecOps: you have added security scanning, secrets management or policy checks to a real pipeline, and supported an audit or assessment in healthcare, finance or another regulated or security-sensitive business.
- Sound incident judgement, clear writing, mentoring, and the ability to explain risk and cost trade-offs to the

CTO.

Experience limited to managed public-cloud services is not sufficient for this role.

Preferred experience

- Healthcare infrastructure, DICOM/PACS or medical-imaging archives, and U.S. healthcare customers.
- Hands-on HIPAA controls or SOC 2 audit support.
- Serving ML models or GPU workloads on self-managed hardware.
- Observability stacks such as Prometheus, Grafana and Loki.

Familiarity with every tool in our stack is not required.

Working arrangement

- Full-time, remote within India, with agreed overlap with U.S. business hours.
- Scheduled maintenance and critical incidents follow a defined escalation and backup rota. Personal availability around the clock is not expected.

📌 AI Infrastructure Engineer (India)
🏢 Natoe.ai
📍 India