Senior Machine Learning Engineer

Recrew AI; Gurugram District; On-site; INR 4,000,000 - 6,000,000; Full time

Job summary

Recrew AI is seeking a Senior ML Engineer to own the full lifecycle of the matchmaking and personalisation engine. You will design and deploy real-time ranking and recommendation infrastructure that personalises user experiences even with sparse data, collaborating with Data, Product and Backend teams to ship scalable ML solutions.

The role emphasizes building embedding systems, graph-based scoring, and cold-start strategies, ensuring robust production deployments and measurable impact on user

Qualifications

• 2–5 years of hands-on experience in personalization, recommender systems, ranking, or search at scale.
• Experience in B2C product startups; social/e-commerce/dating/video platforms.
• Knowledge of collaborative filtering, two-tower retrieval, learning-to-rank, embeddings for ANN search.
• End-to-end ML pipelines from training to production deployment.
• Experience with offline/online evaluation and A/B testing.

Responsibilities

• Design, build, and own end-to-end matchmaking and recommendation systems.
• Develop a real-time adaptive matchmaking engine learning from user interactions.
• Build user embedding systems and graph-based match scoring models.
• Architect cold-start solutions for new users with sparse data.
• Deploy models to production with monitoring and versioning.
• Define offline/online evaluation frameworks and AB tests.
• Partner with Data, Product, and Backend teams to translate business goals into ML solutions.

Skills

Personalisation

Recommendation systems

Ranking

Search systems

End-to-end ML pipelines

A/B testing

Collaborative filtering

Tools

Vector search

Graph-based algorithms

Job description

Job Description:

Role: Senior ML Engineer - Recommendations & Personalisation

Function: Machine Learning / Data Science

Location: Gurugram

Type: Full-time

Industry: Internet / Consumer Tech / Dating & Relationships

About Company

A venture-backed, stealth-stage startup reimagining how people connect. The company is building a next-generation matchmaking platform where AI, community, and content come together.

At its core is a real-time ML recommendation engine delivering deeply personalised experiences - even in sparse data scenarios. Backed by marquee VCs in India and the US, the company has raised $10M+ at seed stage.

The team of 23 operates with the urgency and autonomy of a founding-stage environment. Founders bring extensive experience launching and scaling dating apps and global consumer internet businesses.

Position Overview

This role owns the full lifecycle of the company's matchmaking and personalisation engine - from architecture through production deployment. You will build the real-time ranking and recommendation infrastructure that makes every user's experience feel individually curated, partnering closely with Data, Product, and Backend teams to ship systems that directly define the product.

Role & Responsibilities

• Design, build, and own end-to-end matchmaking, recommendation, ranking, and personalisation systems
• Develop a real-time adaptive matchmaking engine that learns continuously from user interactions and behavioural signals
• Build user embedding systems, similarity models, and graph-based match scoring frameworks
• Architect and implement cold-start solutions to deliver quality recommendations for new users with sparse data
• Deploy models to production with fast iteration loops, model registries, and observability tooling
• Define and track offline and online evaluation frameworks, A/B tests, and metric alignment across the recommendation stack
• Partner with Data, Product, and Backend teams to translate business goals into scalable ML solutions

Must Have Criteria

• 2-5 years of hands-on experience building personalisation, recommendation, ranking, or search systems at scale
• Prior experience at a B2C product company or startup - social, e-commerce, fashion, gaming, dating, or video platforms
• Working knowledge of collaborative filtering, two-tower deep retrieval models, learning-to-rank, and embedding-based ANN search
• End-to-end ML pipeline ownership - both training models and deploying them to production
• Familiarity with offline and online evaluation methodologies, A/B testing, and metric alignment

Nice to Have

• Experience with vector search, graph-based algorithms, or LLM-based approaches for sparse data personalisation
• Prior ownership or leadership of a recsys, feed ranking, or search initiative
• Exposure to cold-start problem-solving in low-data or new-user scenarios
• Background at a high-scale B2C product or consumer startup

What We Offer

• Direct ownership of systems that are core to the product - not peripheral tooling
• Autonomy to shape ML architecture and infrastructure from the ground up
• Collaborative, experimentation-driven culture with responsive, growth-oriented founders