“How is the business performing each day?”
ConditionsNobody has defined which order statuses count as revenue.
RequirementsAgreed sales metrics, orders, AOV, cancellations, and payment performance.
Live bootcamp
Full Stack Analytics Engineer
From raw data to AI
Stop learning Analytics Engineering as a collection of disconnected tools. Build and operate the whole system.
- 3 modules
- 6 weeks
- Weekly office hours
The rise of Full-Stack Analytics Engineering
The role of Analytics Engineer now extends beyond SQL and dbt.
A Full-Stack Analytics Engineer understands how the entire analytics system works—from raw data and infrastructure to trusted analytics, self-service, and AI.
That doesn’t mean owning every component alone. It means understanding how the layers connect, making better decisions across their boundaries, and building systems that remain reliable as they grow.
-
01InfrastructureA stable foundation
-
02TransformationModeled and tested
-
03Trusted analyticsConsistent answers
-
04Self-serviceTeams move faster
-
05AIBusiness-ready intelligence
Learn through realistic business questions
Six weeks of real business requests. One system that grows with them.
Join a simulated ecommerce company as its Analytics Engineer. Work with continuously changing operational data and turn ambiguous stakeholder questions into trusted analytics.
Business requests
“Why do our revenue numbers disagree?”
ConditionsOrder value must be reconciled from line items, discounts, and status rules.
RequirementsA tested order model and an explicit, documented metric definition.
“Which products are driving demand?”
ConditionsProduct prices require validation and inventory is only a current snapshot.
RequirementsProduct and category rankings, discount analysis, and low-stock demand flags.
“Where do customers drop out before ordering?”
ConditionsThere is no session ID, and the order table is the conversion endpoint.
RequirementsAn honest user-level funnel from signup through checkout and purchase.
“Where are our strongest customers and markets?”
ConditionsThe data supports US states—not global performance or true retention.
RequirementsState-level performance and repeat-customer value the business can trust.
“Can teams explore these questions without waiting?”
ConditionsSelf-service and AI are only useful when the definitions underneath are trusted.
RequirementsGoverned metrics, self-service analytics, and a grounded AI experience.
The flagship project
A production-style analytics platform.
Build and operate the whole system—from continuously changing source data to trusted analytics, self-service, and AI.
The six-week journey
Make it work.
Make it right.
Make it scale.
Build one platform in three production stages.
1Weeks 1–2
Launch the platform
The company needs analytics.
- Build the foundation
- Automate the pipeline
- Ship the first dashboard
Outcome
Working infrastructure
2Weeks 3–4
Build trust
The platform needs to become trustworthy.
- Model the data
- Add tests and documentation
- Build reliable workflows
Outcome
Trusted analytics
3Weeks 5–6
Scale delivery
More teams need reliable answers.
- Productionize delivery
- Add observability and semantics
- Enable self-service and AI
Outcome
Scalable, AI-enabled system
Learning platform
Build, submit, get feedback, improve.
Course content, labs, capstones, progress, feedback, and the cohort live together in one platform.
- Hands-on labs tied to the company project
- Automated grading with AI feedback
- Selected human review and office-hour discussion
- Cohort progress and community accountability
LAB 2 · DATA PLATFORM
Build the Snowflake foundation.
The company needs a home for analytics. Set up a Snowflake environment and prepare the foundation for the pipelines you’ll build next.
Environment verifiedAI feedback is ready
Who it is for
You know part of the stack. Now connect it.
Choose the starting point that sounds most like you.
Data
Analyst
You already know:
SQL, dashboards, and metrics.
Add next:
Infrastructure, dbt, engineering practices, and AI.
Junior Analytics Engineer
You already know:
Transformations and pieces of the stack.
Add next:
A complete architecture, production practices, and ownership.
Middle Analytics Engineer
You already know:
Production analytics, dbt, and data modeling.
Add next:
Observability, CI/CD, self-service, and AI agents.
Good fit if…
Comfortable with SQLBasic Git
This is not an absolute-beginner or passive video course.Learn with practitioners
Production judgment—not just tool tutorials.
Learn the architecture, tradeoffs, and failure modes behind modern analytics platforms from people actively doing the work.
Live sessions focus on the decisions that age better than step-by-step recordings: what to build, what to avoid, and how to know when the system is working.
Lead instructor
Oleg Agapov
Analytics engineer and educator with 12+ years across analytics, BI, and data engineering. Oleg has built analytics stacks, led dbt modernization work, and developed AI-assisted engineering workflows for production teams.
- 12+ years in data
- Published AE interview book
- 17.5K+ LinkedIn community
Guest practitioner sessions
A different production perspective every two weeks
Practitioners unpack a real decision from their work—what they chose, what they rejected, and what they learned.
- InfrastructureSystems and scale
- Data modelingTrust and tradeoffs
- AI workflowsPractice and limits
Speaker lineup will be announced before enrollment opens.
Free 2027 roadmap
See where your Analytics Engineering skills need to go next.
Map what you know, spot the gaps between SQL and full-stack ownership, and choose what to learn next.
Full-Stack Analytics EngineerRoadmap
2027
2027
- 01 · FoundationsSQL · Modeling · Git
- 02 · BuildInfrastructure · Ingestion · dbt
- 03 · OperateTesting · CI/CD · Observability
- 04 · EnableSelf-service · Semantics · AI
Inside the roadmap
- Nine capability areas, from SQL to AI analytics
- A progression from analyst to full-stack ownership
- Practical skills and technologies for every stage
- A suggested learning sequence for 2027
Founding cohort · Limited to 25
Build the system—not another throwaway tutorial.
Spend six weeks operating one production-style analytics platform, from raw data to trusted analytics and AI.
Your cohort includes
- 6 live lecturesArchitecture, demonstrations, and production tradeoffs
- 6 live office hoursBring questions, decisions, and work in progress
- Hands-on labs and capstoneBuild against data that changes every hour
- Grading and feedbackAutomated checks, AI feedback, and selected human review
- Practitioner sessionsFresh perspectives across the modern analytics stack
- Cohort communityDiscord support, shared deadlines, and accountability
One-time payment · USD
$599$999
Founding members save $400
You get the complete program at the founding price in exchange for candid feedback that helps shape future cohorts.
Get roadmap + launch updatesEnrollment is not open yetDates, schedule, and policies will be published before enrollment.
FAQ
Questions, answered plainly.
Is this for absolute beginners?
No. You should be comfortable with SQL and basic Git. Foundations such as CLI and data modeling are included, but the pace assumes prior analytics experience.
What is the weekly format?
One live lecture and one live office hour each week, with guest practitioners every other lecture. Most practical work happens asynchronously on the learning platform.
What will I build?
A production-style analytics platform using continuously changing data: infrastructure, transformations, tests, documentation, CI/CD, observability, a semantic layer, and an AI analytics agent.
How is my work reviewed?
You submit dbt artifacts for automated grading and AI feedback. Selected submissions receive human review, and useful mistakes may be discussed during office hours.
Why is the founding cohort $599?
The founding cohort is designed to validate the curriculum and platform while producing the student work, feedback, and outcomes needed for future $999 cohorts.
Are dates, timezone, recordings, and refunds confirmed?
Those operational details have not been finalized yet. They will be published before enrollment opens.