Complex Enablement Programs · Track 02
For enterprises upskilling across teams, roles, and time.
Upskilling multiple teams requires more than individual training courses. We design and deliver coordinated Databricks enablement programmes tailored to different roles, skill levels, and business objectives. From initial assessment to practical application, every programme is structured to build lasting capabilities across your organisation.
We've seen this before
Training more people doesn't always mean building more capability.
Most organisations aggregate training events and track completion rates. The result looks complete on paper and changes little on the ground. And one root cause is consistent: the brief - who needs training and what they think they need - is almost always different from what the actual skills assessment reveals.
The pain is
Mismatched skill levels in the room
Booked by team, not by skill level, some already know the material, others are lost. Instructors teach to the middle; nobody learns efficiently.
The pain is
No sequencing logic
Booked in availability order, not dependency order. Advanced pipeline training before Delta Lake; governance before Unity Catalog. Retention collapses.
The pain is
Technical and leadership tracks planned separately
Engineers build capabilities leadership can't invest in; leadership makes decisions engineers can't support.
The pain is
Completion is measured. Capability isn't.
Attendance and completion aren't proxies for capability. Six months on, no one can say what changed.
How we design it
One programme. Multiple learning paths. Shared objectives.
We create structured, capability-building design: a defined outcome, a mapped skills gap, a personalised and sequenced delivery plan, a measurement layer, and a coordination model.
01
Skills Mapping
We start by assessing the current skill levels of each role and persona, then identify the gaps between where teams are today and where they need to be. This often includes a survey after an initial session, giving us a clear starting point for building personalised learning paths.
02
Layered Curriculum Architecture
We structure each programme in layers so that people with different levels of experience can learn effectively while still progressing towards the same organisational goals.
Opening
Databricks-specific foundations, delivered live to establish a common baseline.
Middle
Standard Academy modules, recommended per individual by role and level, self-paced with support.
Closing
Fully custom workshops built around your real environment, tooling, and ways of working.
03
Sequencing & Dependencies
We make sure people build the right foundations before moving into more advanced or role-specific topics. Shared areas such as Spark, Delta Lake and Unity Catalog are covered first, before participants move into dedicated learning tracks. This helps avoid knowledge gaps and ensures everyone has the context they need.
04
Measurement Framework
We define clear capability benchmarks for each persona and skill area at the start of the programme. Progress is then assessed before, during and after the learning journey, so you can see how capabilities have developed and where further support may be needed.
05
Coordination & Delivery Management
For larger programmes involving multiple teams, cohorts and learning tracks, Datapao coordinates the programme from end to end. We manage scheduling, cohort planning, instructor allocation, office hours and progress tracking, with one programme manager overseeing the full engagement.
What a programme looks like
One designed sequence, from preparation to a capstone project.
The structure below reflects a real, delivered model, a cohort-based programme running roughly ten weeks, built around a few live contact days with personalised self-paced learning and office hours in between.
Before kickoff
Preparation Phase
Incubation Day invite and personalised prerequisite guidance (Python, SQL, account setup). Establishes a common baseline and surfaces gaps early.
Week 1
Incubation Day
Live
Databricks platform and foundations - Apache Spark, Delta Lake, Unity Catalog. Interactive and practical; participants leave with a working mental model.
Weeks 2–5
Self-Paced — Personalised Paths
Self-paced
A post-session survey drives a recommended mix of Databricks Academy courses per person, by role and assessed level. Supported by knowledge checks and office hours.
Week 6
Optimisation Workshop
Live
Storage and performance — Delta Lake internals, cost/performance trade-offs. Opens with an integrated FinOps session on how platform and cloud costs are tracked.
Week 8
Ways of Working Workshop
Live
The most client-specific day - local dev (WSL/VSCode), repo structure, branch-based workflow, unit testing (pytest), deployment (Azure DevOps). Participants clone a repo, write a test, and open a pull request.
Weeks 9–10
Pet Project
Capstone
A real-world task drawn from the organisation's own obligations. Apply everything in a production-like setting, completion earns a Datapao badge.
One room, many starting points
The same cohort holds analysts, engineers, juniors, and managers. The design accounts for all of them.
A single cohort brings together people with radically different backgrounds and seniority. Datapao designs for distinct learner profiles rather than teaching to the middle and adapts delivery in real time.
Profile 01
SQL analysts & report developers
Work primarily in SQL. Beyond fundamentals, they receive targeted software-engineering content, git workflows, Python basics, project structure, unit testing.
Profile 02
Management & team leads
Department heads and senior managers, sometimes leading their own teams through it. A dedicated leadership cohort can align decision-makers with the transition.
Profile 03
Business & system analysts
Don't write production code but need to understand and oversee the lifecycle, from design and testing through to deployment.
Profile 04
Junior developers & interns
Recent graduates with academic Python experience who often progress fastest, a deliberate investment in the next generation of data talent.
Training where the work happens
Not a textbook environment. Your platform, your tooling, your patterns.
Standard product training teaches notebook development in a generic environment. In a complex programme, training runs inside your own Databricks workspace, using representative data and your real toolchain so it transfers directly to Monday-morning work.
Data Assets & Inventory
Establishes a clear view of the organization’s key data assets by identifying what data exists, who owns it, how accessible and reliable it is, and where high-potential or underused assets can support priority use cases.
Your engineering toolchain
Where you develop in VSCode, manage code in git, test with pytest, and deploy through Azure DevOps, the programme trains within that exact toolchain, not a simplified proxy.
A genuine build-together partnership
We work side by side with your team throughout the engagement. Your experts bring the business context; ours bring deep Databricks expertise. Together, we build the assets, documentation, and ways of working needed to keep the programme moving.
Measurement
Completion rates are not capability evidence. We measure what actually changed.
Before
Pre-programme baseline
Each persona is assessed against defined capability benchmarks, documented per role and skill domain. The starting point against which all outcomes are measured.
During
Mid-programme check
Assess whether learning is transferring and whether any cohort needs adjustment — fixed before the programme ends, as when a FinOps module was added mid-flight.
After
Capability delta at close
Assessed against the same benchmarks used at baseline. A role-by-role, skill-by-skill record of what changed, delivered to sponsors with recommendations for the next phase.
Upskilling Central Europe's largest bank for the Databricks era.
Professionals trained
150+
Cohorts delivered
10
Per cohort programme
~10 wks
Pilot to organisation-wide
1 → org-wide
The challenge
OTP Bank - Hungary's largest financial institution - launched a major migration to Databricks across multiple business units. Off-the-shelf training couldn't address its reality: proprietary internal libraries, a custom WSL image, strict data-access controls, and enterprise software-engineering standards, plus a workforce ranging from senior SQL analysts to junior developers and strategic-level managers.
The solution
Datapao co-designed a fully custom, cohort-based programme around OTP's actual environment: live incubation and optimisation workshops, personalised Academy paths per participant, a ways-of-working workshop in OTP's own toolchain, office hours throughout, and a capstone drawn from OTP's real reporting obligations, all in OTP's own workspace on synthetic financial data.
Scale
One pilot cohort in 2025 grew to ten additional cohorts for 2026 — ~150 professionals across the Data Science Technical Tribe, Retail Analytics, and Strategy & Analytics.
Readiness
Faster onboarding, consistent coding standards, and confident use of git, testing, and CI/CD in production.
Alignment
A dedicated leadership cohort aligned decision-makers with the platform transition.
Adoption
A repeatable, cohort-based framework OTP continues to run, with feedback-driven improvement each cycle.
