Skip to content
View vishal-labade's full-sized avatar

Block or report vishal-labade

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
vishal-labade/README.md

Vishal Labade

Staff Data Scientist specializing in experimentation systems, causal inference, and AI evaluation frameworks.

Selected Projects

LLM Experimentation Platform - Conversational behavioral metrics, and causal inference to evaluate prompt strategies, temperature, and model scaling.

AI Evals v2 — Behavioral reliability and context-length evaluation for LLM systems
Experimentation Platform — CUPED / DiD / A/B testing decision framework
NYC TLC Forecasting — Demand modeling and product analytics

image

Pinned Loading

  1. llm_exp_platform_v2 llm_exp_platform_v2 Public

    Experimentation framework for LLM systems using simulated users, conversational behavioral metrics, and causal inference to evaluate prompt strategies, temperature, and model scaling.

    Python

  2. ai_evals_v2 ai_evals_v2 Public

    AI Evals v2 is a structured, reproducible LLM evaluation framework that isolates behavioral reliability from memory capacity. It introduces controlled experiment families, a Memory Compliance Score…

    Python

  3. experimental_platform experimental_platform Public

    Experimentation and causal inference platform for product decision systems, implementing A/B testing, CUPED variance reduction, and Difference-in-Differences analysis with reproducible pipelines.

    Jupyter Notebook

  4. nyc_tlc_github nyc_tlc_github Public

    Deep exploratory analysis of NYC TLC trip data to understand demand patterns, zone-level variability, seasonality, and revenue distribution. Conducted structured EDA on spatial heterogeneity, tempo…

    Jupyter Notebook

  5. spark_setup spark_setup Public

    Containerized Spark cluster with Iceberg table format and MinIO S3 storage, designed for local experimentation and analytics workloads. Includes Docker Compose orchestration, version-pinned environ…

    Shell

  6. ai_evals_v1 ai_evals_v1 Public

    Built a reproducible offline LLM evaluation framework for structured, deterministic scoring. Implements exact, regex, JSON-schema validation, NIC tracking, and numeric invention detection.

    Python