SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
inDrive's Geo Search team is building a proprietary search and recommendation stack for place discovery used by tens of millions of customers across multiple countries. The team spans backend, machine learning, mobile, and QA, working to solve the problem of incomplete or missing commercial map data in emerging markets.
You will own ranking, relevance, and recommendation for geo search end-to-end—from training data pipeline through production models. This is a modeling role with significant production responsibility and founding-team scope.
Key responsibilities:
- Own the complete ranking and relevance pipeline: training data construction, model development, and production deployment
- Build label pipelines that convert search sessions and completed rides into trustworthy training data, correcting for position bias and presentation effects
- Train and maintain models that order search results and confidence models that determine when to serve inDrive's own answers versus falling back to external providers
- Develop query understanding for multilingual markets, including cross-script matching, using large language models for offline labeling and distilling into low-latency production models
- Own evaluation end-to-end: offline replay harness, error analysis, and online A/B testing to measure impact
- Monitor and maintain model health as data distributions and geographic coverage evolve
The technical challenge is significant: suggestions must return within tens of milliseconds per keystroke in cities with sparse map data and mixed-script user input. Model selection is evidence-based: gradient-boosted ranking on behavioral data today, transformers and LLMs where they demonstrably improve query understanding and cross-script matching, and neural reranking where failure analysis justifies the latency cost. Impact is measured through offline evaluation and online experiments, with changes reaching millions of users within weeks.
REQUIREMENTS:
Minimum qualifications:
- 5+ years building models that shipped and improved a production metric
- Direct experience with ranking, search relevance, or recommendation systems, including evaluation methodologies
- Expert-level Python and SQL; fluency with gradient boosting; working knowledge of transformers
- Proven track record taking models to production and owning them post-launch
- Ability to work cross-functionally, as search quality work generates data and engineering tasks owned by other teams
Preferred qualifications:
- Depth in geocoding, autocomplete, or place search
- Learning-to-rank in practice: pairwise and listwise objectives, MRR, NDCG, Hit@k, and their failure modes
- Relevance tuning on OpenSearch or Elasticsearch, including analyzers
- Multilingual or cross-script search experience (e.g., Arabic/Arabizi, Urdu/Roman Urdu)
- Distilling language models under latency constraints; experience in mapping or geospatial products in developing markets
