See the roles to explore, the sequence that makes sense and the evidence to build before your next step. Start with one course or enrol in a discounted pathway package.
Start small. Use what you know. Move forward with evidence.
Free Agentic AI Career Guide 36-page PDF · No email required ↗Your background matters too. Adjust the path below.
Choose an experience level, or read each pathway below.
Start with the software basics you need, then build something you can run, change and explain. Your first advantage is a small working system with clear evidence of your contribution.
Titles and seniority depend on demonstrated skills, current responsibilities and the employer. Years of experience are a starting point, not a qualification.
Why this order: Programming → working application → clear explanation → optional credential. There is little value in defending an architecture you have not yet seen run.
Recommended if you are new to programming or Python. Practise functions, JSON, API calls and debugging before agent frameworks.
Your evidence: Run a small API call and explain a failure.Self-paced · Optional foundationExplore this step →See current regional priceBuild the application before worrying about advanced architectures. Connect models, tools, state and a real workflow.
Your evidence: Run a working agent and make one change independently.Live cohort · 10 January 2027 · Prep available nowIncluded in this pathway packageExplore this step →See current regional priceTurn your project experience into answers that show your contribution and reasoning.
Your evidence: Explain the request flow and one failure case.Self-paced · Structure your answersIncluded in this pathway packageExplore this step →See current regional priceUse associate-level NVIDIA preparation to consolidate concepts after practical work. Certification is optional.
Your evidence: Map concepts to your project, then assess exam readiness.Self-paced · Associate certification prepIncluded in this pathway packageExplore this step →See current regional priceDeveloper Bootcamp includes its Python Prep Module. The separate automation course is optional if you want more foundation practice; do not buy it just to duplicate included prep. Absolute beginners should allow extra time to practise coding.
In your first week of practice, aim to make a small API call, inspect the response and explain one error. If you already code, sketch the first bounded agent you want to build.
A suggested exercise, not a promised seven-day outcome. Start from your current skills and take the time you need.What makes the next level tangible: Move towards independent AI feature delivery when you can change your app, test a failure and explain your own work without the tutorial.
3 included courses individually (final payable) See regional total
Individual final payable prices
Agentic AI Developer Bootcamp + AI Interview Playbook + NVIDIA NCA-GENL Exam Prep.
Optional Python is separate. It is not included in this 3-course bundle or the individual comparison above. Developer Bootcamp already includes its Python Prep Module.
Final payable bundle amount, all inclusive. Automatic enrolment after confirmed payment. One full-payment bundle; standalone instalment offers do not apply. Optional extras, future courses and vendor exams are separate. Already own a course? Ask support before buying; credit is not automatic.
Keep the engineering experience you already have. Add the model, retrieval and evaluation decisions that make an AI feature dependable.
Titles and seniority depend on demonstrated skills, current responsibilities and the employer. Years of experience are a starting point, not a qualification.
Why this order: Build → reason about the design → ground it in data → consolidate with professional exam preparation. Your project gives the architecture and certification topics a practical reference point.
Add agent implementation to the APIs, databases and tests you already know. Skip this stage if you can already demonstrate it.
Your evidence: Trace a request from client to model, tool and response.Live cohort · 10 January 2027 · Prep available nowIncluded in this pathway packageExplore this step →See current regional priceLearn to compare model, retrieval and workflow choices once you have a working implementation.
Your evidence: Write a design decision with a simpler alternative.Self-paced · Architecture decisionsIncluded in this pathway packageExplore this step →See current regional priceMake retrieval a system you can investigate, rather than a vector-database demo.
Your evidence: Explain why a retrieved answer is supported or wrong.Self-paced · Retrieval architectureIncluded in this pathway packageExplore this step →See current regional priceConsolidate agent architecture, retrieval and delivery concepts after practical work. This is NCP-AAI exam preparation; plan the exam when your AI/ML and project experience meet NVIDIA’s guidance.
Your evidence: Use practice questions to find gaps, then test those concepts against your project.Professional exam prep · Check official prerequisitesIncluded in this pathway packageExplore this step →See current regional priceBring the work from this path into a structured interview explanation: context, your decision, alternatives and evidence.
Your evidence: Explain one project decision clearly and answer a follow-up question.Self-paced · Structure your answersIncluded in this pathway packageExplore this step →See current regional priceFor your first-week practice, map a familiar service to an AI request flow. Identify one place where model output needs validation and add that check to a small example.
A suggested exercise, not a promised seven-day outcome. Start from your current skills and take the time you need.Required capability for your next band: production ownership. Enterprise Mastery is our recommended next step to practise recovery, evaluation and end-to-end delivery before taking on senior scope. This is a capability milestone in our recommended path, not a compulsory purchase or an employer requirement.
5 included courses individually (final payable) See regional total
Individual final payable prices
Agentic AI Developer Bootcamp + AI Architect System Design + Production RAG System Design + AI Interview Playbook + NVIDIA NCP-AAI Exam Prep.
Five courses, one package. The five steps above include AI Interview Playbook to help explain your project decisions. NCP-AAI preparation is included; vendor exam fees are separate.
Final payable bundle amount, all inclusive. Automatic enrolment after confirmed payment. One full-payment bundle; standalone instalment offers do not apply. Optional extras, future courses and vendor exams are separate. Already own a course? Ask support before buying; credit is not automatic.
Move from completing features to owning the quality of a component or service. Use design reasoning to guide the build, then investigate operations and reliability in depth.
Titles and seniority depend on demonstrated skills, current responsibilities and the employer. Years of experience are a starting point, not a qualification.
Why this order: Design judgement → integrated build → operational visibility → evaluation. See what the system is doing, investigate failures and make a release decision using evidence.
Frame the boundaries, constraints and alternatives before adding another framework.
Your evidence: Defend the chosen design against a simpler option.Self-paced · Architecture decisionsIncluded in this pathway packageExplore this step →See current regional priceBuild one production-style system across the interfaces a senior engineer must own.
Your evidence: Show failure handling, state and deployment decisions.Live cohort · 7 November 2026 · Pay-in-2 pilotIncluded in this pathway packageExplore this step →See current regional priceTrace your AI workflow, investigate production behaviour and connect failures to dashboards and actionable alerts.
Your evidence: Use a trace and operational dashboard to explain a failing run.Pre-release · Full course by 8 November 2026Included in this pathway packageExplore this step →See current regional priceMake release decisions using repeatable tests and failure categories.
Your evidence: Create a regression set and a release-readiness note.Self-paced · Test and defend qualityIncluded in this pathway packageExplore this step →See current regional priceBring the work from this path into a structured interview explanation: context, your decision, alternatives and evidence.
Your evidence: Explain one project decision clearly and answer a follow-up question.Self-paced · Structure your answersIncluded in this pathway packageExplore this step →See current regional priceUse your first week to write a one-page architecture decision for a familiar workflow: baseline, agent boundary, retrieval choice, failure mode and a quality check. Bring it to your next design discussion.
A suggested exercise, not a promised seven-day outcome. Start from your current skills and take the time you need.What makes the next level tangible: You are ready to explore wider ownership when another engineer can run your component, reproduce its failures and understand the decisions from your documentation.
5 included courses individually (final payable) See regional total
Individual final payable prices
AI Architect System Design + Enterprise Mastery Bootcamp + Agentic Ops Bootcamp + AI Evals & Reliability + AI Interview Playbook.
Final payable bundle amount, all inclusive. Automatic enrolment after confirmed payment. One full-payment bundle; standalone instalment offers do not apply. Optional extras, future courses and vendor exams are separate. Already own a course? Ask support before buying; credit is not automatic.
Start with AI Architect System Design. Your experience is most useful when you can turn constraints into choices, test those choices and guide a system through delivery.
Titles and seniority depend on demonstrated skills, current responsibilities and the employer. Years of experience are a starting point, not a qualification.
For a management route, build on your existing team leadership. Use architecture, evaluation and operations to guide scope, quality and delivery decisions.
Why this order: Architecture → integrated delivery → release judgement → operational ownership. Build the system, decide whether it is ready and show how another team can run it. Worker2Builder adds experience-based positioning and role-focused practice after the technical sequence.
Start here: connect AI choices to the constraints you already encounter in teams and delivery.
Your evidence: Compare a workflow, a single agent and a multi-agent design.Self-paced · Architecture decisionsIncluded in this pathway packageExplore this step →See current regional priceValidate architectural judgement against an end-to-end production-style build.
Your evidence: Connect your design to working code and operational choices.Live cohort · 7 November 2026 · Pay-in-2 pilotIncluded in this pathway packageExplore this step →See current regional priceUse evidence to lead release and risk discussions, beyond a successful demo.
Your evidence: Define acceptance criteria and a regression review.Self-paced · Test and defend qualityIncluded in this pathway packageExplore this step →See current regional priceTurn architectural intent into operational ownership: trace behaviour, interpret evaluation and choose actionable monitoring.
Your evidence: Create an operational dashboard and explain the response to a failing run.Pre-release · Full course by 8 November 2026Included in this pathway packageExplore this step →See current regional priceTranslate your architecture and delivery experience into a role-specific plan. Follow weekly modules and exercises, with role-focused Q&A every two weeks.
Your evidence: Create a target-role map and support your positioning with project evidence.Self-paced · Weekly release · Q&A every two weeksIncluded in this pathway packageExplore this step →See current regional priceFirst-week practice target: create a model-and-retrieval decision matrix for one real use case. Use it to explain your recommendation in a team discussion, with assumptions and a testable next step.
A suggested exercise, not a promised seven-day outcome. Start from your current skills and take the time you need.What makes the next level tangible: Aim for broader scope when you can connect business constraints, implementation, evaluation and support ownership. FDE roles also require hands-on delivery and customer discovery.
5 included courses individually (final payable) See regional total
Individual final payable prices
AI Architect System Design + Enterprise Mastery Bootcamp + AI Evals & Reliability + Agentic Ops Bootcamp + Worker2Builder Career Transition.
Worker2Builder is included. The fifth step builds role positioning around your experience through weekly modules and exercises, with role-focused Q&A every two weeks.
Final payable bundle amount, all inclusive. Automatic enrolment after confirmed payment. One full-payment bundle; standalone instalment offers do not apply. Optional extras, future courses and vendor exams are separate. Already own a course? Ask support before buying; credit is not automatic.
Build the AI judgement to shape technical direction, guide teams and challenge investment decisions. Start with architecture, calibrate it against a real system and quality evidence, then position your experience for an architecture or management role.
Titles and seniority depend on demonstrated skills, current responsibilities and the employer. Years of experience are a starting point, not a qualification.
Director and Head roles are longer-term targets for learners who can combine AI judgement with people leadership, cross-team delivery and business accountability.
Why this order: Architecture → practical calibration → release judgement → role positioning. Operational visibility and a personal Clarity & Direction session help turn those decisions into a focused plan for your next responsibility.
Build the vocabulary and trade-off framework to review an AI proposal with substance.
Your evidence: Challenge a design using constraints, alternatives and evidence.Self-paced · Architecture decisionsIncluded in this pathway packageExplore this step →See current regional priceUnderstand the actual delivery beneath the diagram. Implement or deeply inspect one complete system.
Your evidence: Explain a tool trace, failure boundary and ownership model.Live cohort · 7 November 2026 · Pay-in-2 pilotIncluded in this pathway packageExplore this step →See current regional priceLearn what a responsible release needs and how to question unsupported quality claims.
Your evidence: Write acceptance criteria and an evaluation review.Self-paced · Test and defend qualityIncluded in this pathway packageExplore this step →See current regional priceTranslate your domain, architecture or leadership experience into a role-specific story and plan.
Your evidence: Create a target-role map and evidence-backed positioning.Self-paced · Weekly release · Q&A every two weeksIncluded in this pathway packageExplore this step →See current regional priceConnect leadership decisions to operational evidence: traces, dashboards, alerts and ownership when a workflow fails.
Your evidence: Review a failing run and identify the team action it calls for.Pre-release · Full course by 8 November 2026Included in this pathway packageExplore this step →See current regional priceUse one personal 45-minute AI Clarity & Direction session to prioritise the gaps that matter for your next responsibility. Arrange a mutually available time with support after purchase.
Your evidence: Leave with prioritised next steps tied to your background and target role.One personal 45-minute session · Book with supportIncluded in this pathway packageExplore this step →See current regional priceFor your first week, draft a one-page decision brief for an AI use case: user problem, baseline, investment question, risk owner and acceptance test. Use it to make your next stakeholder conversation specific.
A suggested exercise, not a promised seven-day outcome. Start from your current skills and take the time you need.What makes the next level tangible: Progress becomes tangible when you can prioritise a use case, challenge its design, assign ownership and review evidence of delivery. Hands-on architect and FDE roles still need coding depth.
6 included programmes & sessions individually (final payable) See regional total
Individual final payable prices
AI Architect System Design + Enterprise Mastery Bootcamp + AI Evals & Reliability + Worker2Builder Career Transition + Agentic Ops Bootcamp + AI Clarity & Direction Session.
Six inclusions, shaped around your next responsibility. All six steps above are included: the core programmes, Agentic Ops and one personal 45-minute AI Clarity & Direction session. Contact support after purchase to schedule your session; a bundle purchase does not reserve a meeting time. This is separate from Founder Track’s five-session benefit.
Final payable bundle amount, all inclusive. Automatic enrolment after confirmed payment. One full-payment bundle; standalone instalment offers do not apply. Optional extras, future courses and vendor exams are separate. Already own a course? Ask support before buying; credit is not automatic.
Experience bands are a guide, not a gate. New to coding, returning to tech, or coming from a domain role? Adapt your starting point →
Use a target role and a small piece of evidence to choose your next course. If you already have the skill, move forward. If a foundation is missing, strengthen it first.
Keep your service design, coding and debugging experience. Skip introductory building only if you can already trace a model call, validate a tool request and show an agent failure case. Start with architecture when that foundation is solid.
Start with Python essentials and API practice before the Developer Bootcamp. Its Python Prep Module is included; the separate Python automation course is optional for broader practice. Allow extra time if coding itself is new.
Use your current strengths: QA → evaluation, cloud → operations, domain work → workflow design. Senior technical and FDE roles still need hands-on engineering. A leadership route needs the ability to review the system and its evidence.
Open the first recommended course in your path. Check its prerequisites, format, cohort date and regional price. Buy only the step you need now; come back when you have the evidence to move on.
Browse individual courses →Each experience path has a named package with exact inclusions. Pay securely through Payrouter. After confirmed payment, the bundle membership and its included courses are enrolled automatically. Already own some courses? Ask support before buying; previous payments are not deducted automatically.
Compare your pathway options ↑Compare Diamond if you want broader access and future standard courses under its terms. Builder access and the optional advanced upgrade unlock different parts. Pathway bundles are specific course packages, not Diamond membership.
Compare Diamond access →Bundle checkout uses the regional price shown, including applicable tax, as one full-payment purchase. Included memberships are provisioned automatically after confirmed payment; course dates and access terms apply. Optional courses, Diamond future-course access and vendor exam fees are separate. Previous purchases and coupons are not automatically credited or stacked.
Run a working request. Change a requirement. Demonstrate an error. Explain which parts you implemented and which parts came from a framework.
Compare your approach with a simpler baseline. Explain retrieval, cost, latency, access and failure handling using evidence from the same system.
Let someone else run it from your instructions. Show a regression set, a release decision and a recovery plan. Connect the result to a real user or business need.
First-week targets are suggested practice tasks, not promises of confidence, proficiency, promotion or employment by a deadline. Progress depends on your starting point and practice.
No. Choose a role and identify your next gap. The sequence explains how the courses fit together; it is not a compulsory shopping list. Skip skills you can demonstrate and add specialist courses when the role needs them.
It means the next level of responsibility needs stronger evidence, such as production ownership. Enterprise Mastery is our recommended way to practise that capability. It is not a certification mandated by an employer, and buying a course does not establish a job grade.
Build practical understanding first. NCA-GENL is an associate-level preparation option in the early path. NCP-AAI is professional preparation for learners with relevant depth; check NVIDIA’s current prerequisites. Vendor registration, exam fees and passing the exam are separate from our prep courses.
Check the individual course page. Developer Bootcamp starts 10 January 2027 and includes prep access now. Enterprise Mastery starts 7 November 2026. Self-paced courses can be studied around your schedule. Agentic Ops is pre-release, with the full course due by 8 November 2026.
Worker2Builder is self-paced with a new module and exercise each week and role-focused Q&A every two weeks. The learner community shares session details. Not every module is promised to be available on your enrolment day.
No. These are roles to explore based on the responsibility you want and evidence you can show. Check actual job descriptions. An architect may need hands-on coding; a delivery lead needs more than a course certificate. Your current background matters as much as the year range.
Manifold AI Learning provides the sequence, practice and guidance. You bring the context, do the work and build the evidence. Systems ship. Demos don’t.
Our course sequences and experience bands are Manifold recommendations. For vendor requirements, consult NVIDIA’s official NCP-AAI guidance. They are not a universal hiring framework or a promise of a role. Reviewed 22 October 2026.
