ML Engineer

Most dating apps hand you a stack of profiles and ask you to decide alone. MeantToBe is built on a different premise: that a social community, acting together, generates richer signals about who you are and who you might connect with than any individual ever could.

The product turns that premise into a platform where community participation is both the experience and the data engine. Users engage with each other's potential matches, and every interaction becomes a signal that sharpens personalization for everyone.

MeantToBe is venture-backed by marquee investors in India and the US, and is building at the founding stage.

As ML Engineer at MeantToBe, you will own the design, development, and deployment of the recommendation and personalization systems at the heart of the matchmaking experience.

You will build the matching engine, ranking algorithms, and embedding frameworks that make every user's feed feel curated. The role asks for sharp analytical thinking, the ability to translate product problems into ML problems, and comfort finding signal in noisy, ambiguous data.

You will work closely with the Principal ML Engineer and partner with Data, Product, and Backend teams.

You own the full lifecycle of the recommendation system, from design through production.

• The matchmaking recommendation engine. Build and iterate the real-time adaptive system that learns from user and community interactions.
• Ranking and personalization algorithms. Design the ranking logic that makes each user's feed feel individually curated.
• User embedding and similarity frameworks. Build embedding systems, similarity models, and graph-based match scoring pipelines.
• Cold-start solutions. Explore and integrate approaches that deliver quality recommendations even in sparse data conditions.
• Production ML infrastructure. Deploy models using fast iteration loops, model registries, and observability tooling.
• Cross-functional delivery. Partner with Data, Product, and Backend to ship experiences that work end to end.

• Built recommendation systems in a B2C product. You have worked on personalization, feed ranking, or search at a consumer platform, social app, e-commerce, or video product.
• Translated product problems into ML problems. You can take an ambiguous business question, frame it as a data science problem, and design a solution.
• Found signal in noisy or sparse data. You have worked with incomplete or messy data and made defensible modeling calls without waiting for clean inputs.
• Shipped models end to end. You have taken a model from design through training to a production system, even if not every step every time.
• Evaluated models rigorously. You understand offline and online evaluation, A/B testing, and how to align metrics to real product outcomes.
• Worked across the recommendation technique stack. You have exposure to collaborative filtering, two-tower retrieval, learning-to-rank, embeddings with ANN search, or LLM-based personalization approaches.

• Immerse before you build. The first weeks are for understanding the product, the data, and the user behavior; intuition built here shapes everything after.
• Own the problem, not just the task. You identify what is worth building next, not just execute a backlog handed to you.
• Think in signals. The best work here starts with a question about the data, not a solution looking for a problem.
• Move fast with judgment. The stage rewards quick learning loops; ship a working version, measure it, and iterate.
• Collaborate across functions. Product, Data, and Backend are close partners; good ML here is built with them, not handed to them.
• Show up in person. The team works from Gurgaon; presence and engagement in the early months are how trust and momentum are built.

Cash compensation of ₹33.8L to ₹45L a year, with flexibility for exceptional candidates. Meaningful ESOPs are available and structured case by case based on seniority and fit. You report to the Principal ML Engineer and own the core recommendation system. The role gives you full ownership of one of the hardest personalization problems in Indian consumer tech.