About the Role
You'll own the online systems that take a catalogue change to a ranked result in front of a shopper, encompassing search, retrieval, and recommendation serving. This includes managing the streaming data that ensures they remain fresh and the experiments that facilitate deployments. You will be a vital part of the company's AI team, which is crafting a self-owned discovery stack across our marketplace and merchant stores that features hybrid lexical and vector search, Arabic-first query understanding, and learned ranking shared between search and recommendations. While data scientists focus on relevance and ranking models, you will oversee the platform to make these models fast, fresh, reliable, and shippable.
The work is divided roughly into half online serving, one third streaming data, and the remainder into integration with product surfaces. You will be positioned on the ML Platform team, working closely with search data scientists and the ranking lead across both search and serving for our recommendation systems.
Ready to apply for roles like this?
Unlock the company name and direct application link. Subscribers get instant access to fresh jobs across Dubai, Abu Dhabi and Riyadh, many with visa support.
Unlock employer & apply directly
What You'll Do
Online Serving
- Build and run query and recommendation serving, including retrieval, score fusion, filtering, and caching.
- Maintain tight latency budgets, targeting double-digit-millisecond p95 for search-as-you-type.
- Design graceful degradation to ensure serving maintains functionality when a model or vector path is slow or down.
- Develop experiment plumbing involving shadow traffic, A/B assignment, guardrail metrics, and rollback processes.
Streaming Data
- Construct change-data-capture from source systems into search indexes and feature stores, monitoring for freshness.
- Ensure embeddings remain fresh for both new and updated items.
- Execute zero-downtime reindexing of multi-million-document collections.
- Develop feature logging that translates production traffic into training data for ranking models.
Integration
- Deliver changes to product teams' backends and clients, including new endpoints, cutovers, and experiment flags.
- Collaborate with product engineering teams on reviews, release windows, and production ownership.
Operability, Throughout
- Implement monitoring, alerting, runbooks, and manage your share of on-call responsibilities for what you ship.
Requirements
- 5+ years of experience building backend or platform systems in production.
- 2+ years of experience operating a production search engine (such as Elasticsearch, OpenSearch, Solr, Typesense, Vespa, or similar), with expertise in index lifecycle, analyzers, reindexing at scale, and debugging relevance issues alongside data scientists.
- Experience with low-latency services under real traffic and the ability to articulate reasons for deviations in p95 metrics.
- Proficiency in streaming ingestion technologies such as Kafka, Debezium, or similar change-data-capture mechanisms, along with idempotent backfills.
- Strong programming skills in Python and comfort with deploying in TypeScript.
- Experience shipping changes through A/B testing and staged rollouts.
- Familiarity with agentic coding tools like Claude Code, with expectations for one engineer plus agents to accomplish what previously required a team.
