What you will do
• Improve the quality of search results, so that the most relevant jobs appear at the top.
• Work out what a user actually means when they type a search — the role, the locality, the experience level, the pay they expect — even when the query is a few words typed in Hindi, Hinglish or a regional language, with spelling that follows how the word sounds.
• Make distance a first-class part of ranking. For most of our roles, a job close to home is a fundamentally better result than a well-matched one across the city.
• Keep results genuine and current — filled or low-quality listings ranked highly are a bad user experience, so quality and freshness are part of the model, not an afterthought.
• Build and improve recommendation models that suggest relevant jobs to jobseekers and relevant candidates to recruiters.
• Decide how "good" is measured, and prove your changes helped by running experiments on live traffic.
• Work with very large volumes of user activity data to build the inputs your models need.
• Partner with product managers and engineers to take models live, and present results clearly to both technical and non-technical audiences.

What we are looking for
• 4–6 years of hands-on machine learning experience.
• At least 2 of those years spent on search result ranking, recommendations,



or personalisation — for a live product with real users. Academic projects, competitions, and internal prototypes do not count towards this.
• Strong programming in Python and robust SQL.
• Experience handling very large datasets using tools such as Spark, PySpark, or Hive.
• Practical experience with the machine learning techniques used for ranking and recommendations.
• Experience designing and running A/B tests, and the ability to say clearly what your change did to a business number.
• Experience working with text data — matching what a user is looking for against what a job description says. Experience with Indian-language or transliterated text is a strong plus.
• The ability to explain a technical trade-off in simple language to a product manager or business stakeholder.

Nice to have
• Experience with search technologies such as Elasticsearch, Solr, or OpenSearch, or with large-scale similarity search.
• Experience with deep learning approaches to recommendations and retrieval.
• Experience in a two-sided marketplace, where both sides of the platform have to be kept happy.
• Practical use of large language models to support search and relevance work.
• Experience with production ML tooling such as Airflow, MLflow, or feature stores.

📌 Lead Engineer - AI/ML (Noida)
🏢 Info Edge
📍 Noida