Simachew Mengiste
Researcher and data scientist, building analytical systems.
I work across computational neuroscience, time-series analysis, NLP tooling, and data products, with an emphasis on reproducible pipelines and practical decision support.
Mathematics · Machine Learning · Systems
Selected Projects
A few representative pieces of work spanning scientific modeling, applied analytics, and practical language tooling.
tNet Analysis
(temporal graphs)
2024
Modular toolkit for constructing, simulating, and analyzing temporal networks, with integration and segregation metrics.
Temporal graphsDynamic functional connectivityNull modelsRandom walks
Market Analytics
(LLM-assisted pipeline)
2025
Reproducible time-series analytics pipeline for commodity price data, from ingestion and feature engineering to modeling and evaluation.
Time-seriesSnakemakeSQLiteBaseline model
Amharic Normalization & Canonical Representation
(Canonical Orthographic Framework)
2025
Deterministic orthographic normalization framework for Amharic, featuring a reversible canonical representation (CAR), ambiguity-aware Latin decoding, and bidirectional punctuation standardization.
Canonical encodingTransliterationFastAPINLP preprocessing
Focus
Scientific modeling
Computational neuroscience, temporal networks, graph structure, and simulation workflows.
Applied analytics
Data pipelines, baseline models, evaluation, and clear analysis for decision-oriented work.
Usable systems
APIs, dashboards, NLP utilities, and tools that make models easier to inspect and apply.