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 owns the data, ranking algorithms, and measurement infrastructure for place search, autocomplete, geocoding, and location-based personalization.
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-level scope.
Key responsibilities:
- Own the complete ranking and relevance system: build training data pipelines that convert search sessions and completed rides into trustworthy labels, accounting for position bias and presentation effects
- Train and deploy models that order search results and confidence models that decide when to serve inDrive's own answers versus falling back to external providers
- Build 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: build offline replay harnesses, conduct error analysis, and run online A/B tests to measure impact
- Maintain model health post-launch as data distributions and geographic coverage evolve
The technical challenge is significant: suggestions must return within tens of milliseconds per keystroke in markets where commercial map data is incomplete or missing, and users type in mixed scripts. Model selection is evidence-based: gradient-boosted ranking on behavioral data today, transformers and LLMs where they justify latency trade-offs, and neural reranking where failure analysis supports it.
You will work cross-functionally with backend engineers, mobile developers, QA, product managers focused on relevance, and geo analysts. Changes reach millions of customers within weeks.
REQUIREMENTS:
Minimum:
- 5+ years building models that shipped and improved production metrics
- 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 across teams, as search quality work generates dependencies for data and engineering teams
Preferred:
- 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)
- Experience distilling language models under latency constraints
- Background in mapping or geospatial products in developing markets
