Senior Data Scientist (Geo Search)

inDrive
Almaty
Workplace: HybridFull timeFunction: Data Science & Machine LearningExperience: 5+ yearsSkills: ["Cross-functional collaboration","Evidence-based decision making"]

Own ranking, relevance, and recommendation models for geo search end to end. Define relevance measurement, create trustworthy training labels from messy behavioral logs and completed rides, and ship models that serve within tens of milliseconds per keystroke. Drive offline evaluation and online A/B tests, keep models healthy post-launch, and collaborate across backend, ML, mobile, and QA teams to improve conversion and mean reciprocal rank.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
inDrive
inDrive
3 hours ago

Senior Data Scientist (Geo Search)

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 3 hours agoStatus: Live
Reposted: similar role first listed 4 days ago

Job Summary

Own ranking, relevance, and recommendation models for geo search end to end. Define relevance measurement, create trustworthy training labels from messy behavioral logs and completed rides, and ship models that serve within tens of milliseconds per keystroke. Drive offline evaluation and online A/B tests, keep models healthy post-launch, and collaborate across backend, ML, mobile, and QA teams to improve conversion and mean reciprocal rank.
Location: Almaty
Workplace: Hybrid
Employment Type: Full time
Job Function: Data Science & Machine Learning
Seniority: Mid level

Key Responsibilities

  • •Own ranking, relevance, and recommendation for geo search from training data through production serving.
  • •Build the label pipeline from search sessions and completed rides, correcting for position bias and other presentation effects.
  • •Develop and maintain models that order results and a confidence model that decides per query whether to serve internally or fall back.
  • •Create query understanding for multiple markets using LLMs for offline labeling and distill into low-latency models for real-time keystroke paths.
  • •Drive evaluation end to end with offline replay harness, error analysis, and online A/B tests; monitor and keep models healthy after launch.

Pay and Benefits

Perks:Learning ProgramsTalent ExchangeCo-financed Courses

Key Requirements

  • •5+ years building models that shipped and improved a production metric.
  • •Direct experience with ranking, search relevance, or recommendation systems and their evaluation.
  • •Expert Python and SQL, with fluency in gradient boosting and working knowledge of transformers.
  • •Experience taking a model to production and owning it after launch.
  • •Ability to work across teams as search-quality work generates data and engineering tasks.
Experience:5+ yearsSearch relevanceRankingRecommendation systemsProduction ML
Skills:Cross-functional collaborationEvidence-based decision making
Languages:ArabicUrdu
Tech Stack:PythonSQLGradient boostingTransformersLarge language modelsOpenSearchElasticsearch

Company Brief

inDrive
A ride-hailing and mobility platform that lets passengers and drivers negotiate fares directly in real time. It also offers intercity travel, courier, and delivery services in multiple countries.
Industry: Mobility Platforms
Company Size: Enterprise (1,001+ employees)
Growth: Scaleup
Headquarters: Mountain View, United States
Founded: 2013
WebsiteLinkedIn