3. Data, Science & AI

Generates secure admin dashboards for PostgreSQL and MongoDB, enabling schema management, CRUD operations, and customizable security models.

Calculates and tracks key performance indicators (KPIs) from pilot data, enabling benchmarking against objectives and generating quantitative reports and dashboards.

Performs exploratory data analysis on pilot datasets to identify patterns, anomalies, and initial insights for hypothesis generation and data quality validation.

Scopes analytics initiatives by defining key metrics, data sources, and analysis methods to align with business objectives and stakeholder requirements.

Provides tools for testing and evaluating Large Language Models (LLMs) using frameworks like promptfoo, trulens, and evals.

Orchestrates tasks across multiple LLM providers (OpenAI, local, cloud) to optimize cost and latency in AI workflows.

Enforces data quality rules including missing values, schema compliance, and anomalies on data streams and repositories before analysis.

Detects and alerts on statistical outliers and threshold breaches in real-time data streams to trigger governance reviews and corrective actions.

Enables Claude to delegate tasks to Google's Gemini AI, utilizing its 1M token context and Google Search grounding for enhanced AI collaboration.

Fetches and analyzes current weather conditions and forecasts for specified locations, enabling data-driven decisions and Chainlit chat integration.

Generates high-performance scientific GUI applications using PyQt and visualization tools within a data-centric architecture for data analysis.

Transcribes YouTube videos and local audio files to searchable markdown text using AssemblyAI's speech-to-text technology.

Provides programmatic access to Up Banking API for querying accounts, transactions, and generating spending analysis summaries.

Statistical genetics toolkit for fine-mapping, LDSC, and TWAS using GWAS summary statistics, heritability, and genetic correlations, generating publication-ready reports.

Enables automated web data collection via Apify Actors, supporting repeatable and scalable data acquisition pipelines.

Manages linked memory entities and relationships in Neo4j, enabling efficient storage, deduplication, and graph-based retrieval for project continuity.

Generates LINE Bot-compliant images using Gemini AI for backgrounds and matplotlib for programmatic Chinese labels, ensuring under 300KB and mobile readability.

Handles memory-safe data import with streaming, explicit type validation, error logging, and idempotent database writes for robust data pipelines.

Automates extraction of structured meeting notes from transcripts, including timestamps, action items, and highlights for team meetings and workshops.

Classifies biographical stories into lifestyle categories for structured data analysis and organization of personal narratives.

Executes probabilistic Monte Carlo simulations to evaluate trading strategy robustness and risk exposure.

Converts UML class diagrams into Neo4j graph database by extracting entities and relationships via vision models and generating Cypher.

Enables semantic and full-text search over codebases and documentation, returning relevant files with relevance scores.

Provides Django template integration for Chart.js to render interactive statistical graphs with data labels and dynamic data injection.