Beating the averages at JustDial with Julia

on YouTube ↗15m

" summary

JustDial rewrites its pricing, budgeting and search pipelines in Julia, replacing a PHP plus C module stack with a single pricing engine that trains overnight and ships models for live negotiation by 5am. The pipeline ingests roughly 3 TB/day of city-level signals across about 46,000 locations, turns velocity and historical features into per-category price estimates, and delivers large compute and latency savings (roughly a 60% reduction in batch compute) enabling daily automated pricing decisions.

AI-written summary · about

" chapters (3)

1. 0:00Beating the averages at Justdial with Julia
2. 1:40Budgeting Match demand and supply 100,000 categories of products and services 1000s of localized markets Use velocities, historical values of demand and supply Compute price for every category in each geography Pricing packages add one more complexity - Platinum Diamond, Gold and Silver
3. 7:58Rating Engine Solution with Julia, Apache Phoenix & Apache HBase Julia implementation in progress

" related talks

• 14mJulia (workshop)Ranjan Anantharaman, Shashi Gowda · Cypher 2017
• 43mJulia for Data Analysis features, interfaces and future directionsSimon Byrne · PyData London 2016
• 27mJulia (workshop)Ranjan Anantharaman, Shashi Gowda · Cypher 2017
• 32mJulia for PythonistasKelly Shen · PyData New York City 2019
• 01h 27mIntroduction to Julia for Scientific Computing and Data ScienceDavid Higgins, Robert Schwarz · PyData Berlin 2017
• 30mJulia for Modern Data AnalysisJosh Day · PyData Carolinas 2016
• 01h 37mTech Tutorial: Use Julia for your next scientific computing job!P. Szufel · International Data Science Conference 2021
• 01h 38mAn Introduction to JuliaJane Herriman · PyData Los Angeles 2018