Agentic AI Infrastructure Engineer
ANKUR NARANG
DeepTech Entrepreneur, Technology Leader
Kush Khurana
AI & ML Leader, Venture Partner - DCX
.jpeg&w=768&q=75)
AI agents are easy to demo. Building reliable agent infrastructure is not.
As companies move from AI experiments to real agentic systems, the hardest problems are shifting from prompts and prototypes to infrastructure: orchestration, model routing, memory, evaluation, observability, security, reliability, latency and cost.
Most engineers can now build a basic agent. Far fewer know how to make one survive production failures, handle long-running workflows, recover from broken tools, control model spend, trace why an agent failed, or safely connect agents to enterprise systems.
That gap is creating a new engineering discipline: Agentic AI Infrastructure Engineering.
This course teaches you how to build the platform layer behind production agents. You’ll design and implement runtimes, gateways, memory services, evaluation pipelines, AgentOps, security controls and scalable execution patterns.
The goal is simple: move from “I can build an agent” to “I can engineer the infrastructure that lets agentic systems operate reliably at enterprise scale.”
What you’ll learn
Learn to design, deploy, scale, secure, and operate production-grade infrastructure for reliable AI agents and agentic platforms.
Architect control planes, runtimes, gateways, shared services and deployment layers for enterprise agent platforms.
Separate application, control, runtime and data-plane responsibilities.
Create a reusable reference architecture for multi-team agent platforms.
Containerize and deploy agent workloads using cloud-native and Kubernetes-based infrastructure patterns.
Implement autoscaling, resource isolation, health checks and resilient execution.
Support real-time, asynchronous and long-running agent workloads.
Design secure infrastructure for multiple teams, users and workloads on shared agent platforms.
Implement tenant isolation, quotas, identity boundaries and noisy-neighbor protection.
Meter infrastructure usage and allocate platform costs by tenant or team.
Create centralized services to register, configure, deploy and manage large fleets of agents.
Manage versions, policies, configuration, tool access and agent lifecycle centrally.
Implement health controls, rollback, remote configuration and kill switches.
Define SLIs, SLOs, error budgets and reliability targets for agent infrastructure.
Build fleet-wide metrics, logs, traces and operational dashboards.
Design incident response, failure recovery and disaster-recovery strategies.
Model concurrency, throughput, queue depth and infrastructure capacity for agent workloads.
Model concurrency, throughput, queue depth and infrastructure capacity for agent workloads.
Build cost-allocation and FinOps dashboards for enterprise AI infrastructure.
Learn directly from expert instructors
ANKUR NARANG
Three decades building HPC and AI systems processing billions of transactions
.jpeg&w=384&q=75)
Kush Khurana
AI product and deep-tech builder, Visiting Professor at Ashoka University
Madhusudan Kumar
BTech & MTech IIT Mumbai. AI Lead with solid Agentic Experience
Who this course is for
AI / ML / LLM Engineers: Move beyond building agents to designing shared runtime, platform & infrastructure that prod AI systems depend on.
Backend / Cloud / Platform Engineers: Apply cloud and platform engineering skills to build scalable infrastructure for enterprise AI agents.
Senior Engineers / Architects: Design enterprise agent platforms, control planes, multi-tenant runtimes, security and FinOps.
What's included
Live sessions
Learn directly from your instructors in a real-time, interactive format.
Lifetime access
Go back to course content and recordings whenever you need to.
Community of peers
Stay accountable and share insights with like-minded professionals.
Certificate of completion
Highly reputed co-branded certificate from DeepCoreX Inc and AI CERTs. Share your new skills with your employer or on LinkedIn.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Course syllabus
20 lessons • 4 projects
Week 1
Enterprise Agent Platform Architecture
Week 2
Runtime, Cloud Infrastructure & Multi-Tenancy
Schedule
Live sessions
4 hrs / week
Projects
2 hrs / week
Async content
2 hrs / week
Frequently asked questions
Maven for Teams
Reimbursement
Get your company to pay
Everything L&D needs: email template, receipts, and certificate of completion.
Get reimbursedTeam discount
Learn with your teammates
Save 20%+ when 2 or more teammates enroll in the same cohort.
Save 20%+ with a teamPrivate cohort
Run a cohort for your org
A dedicated cohort with a custom schedule and curriculum, tailored to your team.
Book a private cohort$899
USD