12-Week Live Training · Small Cohorts · Industry-Grade
Intensive, cohort-based tracks for STEM graduates and early-career professionals. Two programmes — one in Data Science & Causal Inference, one in Python & Software Engineering. Both built around how real teams work.
The Approach
Both tracks share the same underlying philosophy: the goal is not to teach software or methods in a vacuum, but to build the judgment to use them correctly.
01
In data science, a badly designed experiment gives worse answers than no experiment. In software, coding before thinking produces code no one can maintain. Both tracks start from this truth.
02
Selection bias, Simpson's paradox, mutable default arguments, bare except clauses. The specific ways things go wrong in practice are a core part of the curriculum — not footnotes.
03
Each week builds on the last. By the end, students are not completing isolated problems — they are applying a coherent body of knowledge to a project they designed themselves.
04
Sessions are capped so that questions get answered, work gets reviewed, and feedback is personal. Not a recorded video with a forum no one reads.
The Two Tracks
The tracks are independent and can be taken separately or sequentially. Together they cover the full stack of a modern data-focused engineering role.
How modern data teams answer "what caused what?" — in product, marketing, and policy settings. Covers A/B testing, causal inference with observational data, and decision-making under uncertainty.
Best suited for: candidates aiming for data scientist, product analyst, or experimentation roles.
From writing scripts to building tested, structured software — the way engineering teams actually work. Covers the full journey from language foundations to APIs, databases, and data analysis.
Best suited for: STEM graduates, bootcamp alumni, and analysts who want to close the gap between "it works on my machine" and professional engineering standards.
Week by Week
Both tracks are structured in four phases — each phase builds a distinct capability before the next one begins. Nothing is introduced before students are ready for it.
Phase 1
Thinking
Weeks 1–4
Week 1
Causality & Decisions
Correlation vs causation · framing business decisions
Week 2
Why Naive Analysis Fails
Selection bias · confounding · Simpson's paradox
Week 3
Introduction to Experiments
Randomisation · treatment vs control · potential outcomes
Week 4
Designing Good Experiments
Hypotheses · metrics · experiment pitfalls
"Why causality is hard"
Phase 2
Measurement
Weeks 5–8
Week 5
Hypothesis Testing
p-values · confidence intervals · misinterpretations
Week 6
Power & Sample Size
Effect size · time vs accuracy tradeoffs
Week 7
Experiment Pitfalls
Multiple testing · peeking · metric hacking
Week 8
Experiments to Decisions
Practical vs statistical significance · business context
"How experiments solve it"
Phase 3
Causal Methods
Weeks 9–11
Week 9
Regression for Causal Inference
Regression as adjustment · assumptions
Week 10
Interrupted Time Series
Trend vs level change · pre/post analysis
Week 11
Synthetic Control
Weighted comparison groups · constructing counterfactuals
"What to do when experiments fail"
Phase 4
Synthesis
Week 12
Week 12
Choosing the Right Method
Decision framework · justify assumptions · present recommendation
Decision Framework
Can randomise → A/B test
Time-based →
ITS
Comparable groups → Synthetic Control
Observational + covariates → Regression
"How to choose"
Phase 1
Language & Thinking
Weeks 1–4
Week 1
Environment & Mental Model
VS Code · variables · types · f-strings
Week 2
Data Structures
Lists · dicts · tuples · mutability
Week 3
Control Flow
Conditionals · loops · comprehensions
Week 4
Functions & Scope
Parameters · returns · scope · recursion
"How Python thinks"
Phase 2
Structure & Scale
Weeks 5–8
Week 5
Objects & Classes
__init__ · methods · inheritance · when not to use classes
Week 6
Modules & Packages
Import system · venv · pip · stdlib tour
Week 7
Files, Errors & I/O
CSV · JSON · try/except · context managers
Week 8
CLI Tools
argparse · stdin/stdout/stderr · exit codes
"Writing code others can use"
Phase 3
Real-World Patterns
Weeks 9–11
Week 9
Testing & Unit Tests
pytest · edge cases · mocking · test-driven habits
Week 10
APIs, Data & FastAPI
requests · REST APIs · FastAPI · pagination
Week 11
Databases & SQL
SQL fundamentals · T-SQL · queries · schemas
"Code that works in practice"
Phase 4
Synthesis
Week 12
Week 12
Capstone & Decision Framework
Design before coding · script vs module vs package · present your choices
Decision Framework
One-off task → script
Shared logic →
module
Distributable tool → package
Related state + behaviour → class
"How to choose"
Engagement Model
Training is delivered in small cohorts to allow for genuine interaction, feedback, and realistic project work.
12-week live programme in a small group. Weekly sessions, structured exercises, and a capstone project. Cohort size is capped to allow genuine interaction and code review.
Everything in the cohort track, plus regular 1:1 sessions for code review, career guidance, and personalised feedback on your project work. The most direct route to professional-level output.
Most popularDelivered to a specific team or organisation. Scope, pace, and examples are adapted to your stack and use case. Suitable for teams onboarding junior engineers or upskilling analysts.
Why This
Each week builds on the last. By the end, students present a complete project they designed themselves — not a portfolio of isolated exercises.
The specific ways things go wrong in practice — Simpson's paradox, bare except clauses, hardcoded paths — are a core part of the curriculum, not footnotes.
No toy frameworks. No shortcuts that need to be unlearned. The tools, workflows, and habits are the ones you will encounter on day one of a real role.
4–10 people. Sessions are capped so that questions get answered, work gets reviewed, and feedback is personal — not a recorded video with a forum.
Both tracks are built around the same principle: the goal is not to memorise tools, but to develop the judgment to know when and how to use them.
Both tracks start from zero and move deliberately. What is required is the willingness to treat this as serious professional training — not casual coursework.
Pricing depends on track, cohort size, and level of mentoring.
Get in touch