← All work
Case StudyTier BResearchSpring 2023
Persian Poetry Generation & Style Transfer
Academic NLP project preparing a large Persian poetry corpus and implementing a generative/style-transfer research prototype.
AIResearchData
Context
Spring 2023 · Academic research prototype · Maturity: Research.
Problem
Explore generation and stylistic modeling across Persian poets.
Users
Academic research context.
My Role
3-person team; full code implementation contribution.
Product Decisions & Responsibilities
- Structured the experiment and data preparation around poet/style context.
Technical Approach
- Prepared approximately 1.46M records across 68 poets and implemented model code.
PythonNLPGenerative AIData preparation
Constraints
- Research prototype with no public launch or reliable user-quality benchmark.
Product Flow
- Collect/prepare poetry
- Label/style context
- Train/experiment
- Inspect generated output
Outcome
- ~1.46M prepared records across 68 poets.
What I Learned
- Data preparation quality is foundational for language-model experiments.
Evidence Boundary
What this does — and does not — prove
No public-launch, traction, or quality benchmark claim.
Next Iteration
Extend the product only after the next decision can be tied to stronger user, operational, or business evidence rather than adding features for their own sake.