Skip to content

Case Studies

This section will collect deeper technical writeups.

For now, it starts as a curated index. It does not invent case studies that have not been written yet.

The goal is to keep this section useful while the deeper writeups are prepared gradually.


Current documented references

RetainAI

RetainAI is the main AI-native product and platform case.

It can support future case studies around:

  • AI-native decision intelligence;
  • multicloud backend architecture;
  • GCP Cloud Run frontend deployment;
  • AWS Lambda backend deployment;
  • security and cost controls;
  • vector/RAG architecture decisions.

Documentation:


VersoVector

VersoVector can support future case studies around:

  • classical ML/NLP workflows;
  • semantic and emotional analysis of poetic language;
  • reproducible notebooks;
  • model interpretation;
  • communicating research-oriented technical work.

Documentation:


Data & Cloud Engineering

This area will contain only sanitized and non-sensitive writeups.

Potential topics include:

  • data pipeline design;
  • cloud data architecture;
  • production troubleshooting;
  • PostgreSQL and BigQuery integration patterns;
  • batch processing and scheduler design;
  • operational safety checks;
  • cost-aware architecture choices.

Current index:


Future case-study candidates

These are candidates, not published deep dives yet:

CandidatePossible focus
RetainAIAI-native product architecture, multicloud deployment, backend cost controls
VersoVectorClassical ML/NLP and poetic-language analysis
MIAD-RAG-RealEstateRetrieval-augmented generation and analytical product thinking
GenderMovieClassificationSupervised learning and text classification practice
Legacy Big Data LabDistributed data processing foundations
Sanitized Data & Cloud Architecture CaseProfessional data platform patterns without sensitive information
AppMarkowitzPortfolio optimization and applied analytics foundations

Publishing rule

A case study should only be added when it has:

  • a clear problem statement;
  • public or sanitized context;
  • architecture or workflow explanation;
  • tradeoffs and decisions;
  • reproducible or explainable artifacts;
  • no sensitive client, company or personal data.

Until then, this page remains an honest index rather than a list of invented essays.

Last updated:

Built with VitePress and deployed with GitHub Pages.