Position: Senior AI/ML Engineer
Experience: 5–8 Years
Location: Delhi
Work Mode: Hybrid – 2 Days WFO
Working Days: 5 Days a Week
Shift Timings: 9:00 AM – 6:00 PM
Notice Period: Immediate to 30 Days Preferred
Interview Process: 3 Rounds | Final Round – Face-to-Face
Key Responsibilities
Data Foundation & Instrumentation
- Design and own clickstream and impression-logging pipelines.
- Define event taxonomies that support downstream ML models.
- Normalize and match products across gift cards, merchandise, and services.
- Build product taxonomies and deduplication frameworks to support co-occurrence analysis.
Search & Retrieval
- Own search quality, including zero-result rate, query understanding, synonym/typo handling, and relevance ranking.
- Build semantic search and eligibility Q&A systems using catalogue and plan-rule data.
- Implement retrieval and generation systems with citations and appropriate abstention when information is uncertain.
Recommendations & Personalization
- Build co-purchase and cohort-based recommendations and collections.
- Develop session-based recommendations and in-session intent models.
- Build personalized collections using browsing and redemption behavior.
- Develop learned ranking models using user, item, cohort, and contextual features.
- Ensure eligibility, balance, and availability are enforced as hard constraints.
- Build utilization intelligence to identify users at risk of forfeiting funds and determine relevant engagement opportunities.
Responsible & Privacy-Aware ML
- Maintain strong architectural boundaries around health-adjacent interaction data.
- Ensure sensitive data is appropriately segregated at the schema level.
- Design systems to minimize confidently incorrect eligibility answers and enable models to abstain when they do not have sufficient information.
Serving & Reliability
- Deploy, serve, and operate ML models in production.
- Own model latency, availability, monitoring, and reliability.
Experimentation & Business Impact
- Build robust offline evaluation frameworks that reflect real-world outcomes.
- Conduct A/B testing against business metrics such as utilization rate, order frequency, and search success.
- Measure and communicate the business impact of ML initiatives.
Must-Have Skills
- 5+ years of experience in production ML or ML-adjacent engineering.
- Strong expertise in either Search & Ranking or Recommendations & Personalization, with practical understanding of the other.
- Proven experience building and shipping a production ML system from the ground up.
- Strong expertise in embeddings, vector search, hybrid lexical + semantic retrieval, and reranking.
- Experience with vector search technologies such as FAISS, HNSW, pgvector, or equivalent managed platforms.
- Experience with learned ranking, including GBDT rankers and neural/sequential approaches.
- Expert-level Python and strong SQL skills.
- Experience with Spark or equivalent data-processing frameworks.
- Experience with Airflow, Dagster, or equivalent workflow orchestration tools.
- Hands-on experience deploying and operating ML systems in production.
- Strong understanding of model serving, monitoring, reliability, and scalability.
- Rigorous approach to ML evaluation with the ability to connect technical improvements to measurable business outcomes.
- Strong ownership and self-direction, with the ability to define roadmaps, manage stakeholders, negotiate scope, and deliver independently.
