Applied AI / Data Scientist

What you will do

Own data and AI solutions end to end: understand the problem, analyse the data, pick the approach, then build, test and iterate with users.

Combine data from multiple sources to find patterns, diagnose performance and support decisions.

Use data science, ML or GenAI depending on the problem — not every problem needs the same tool.

Build usable tools such as automated analysis, decision tools, AI workflows and simple apps, connected to databases and APIs.

Turn one-time analyses into repeatable products. Measure the value they create and improve them based on real usage.

Work directly with business and manufacturing teams.

Initial projects

Style Performance & Decision Engine — combine sales, inventory, marketplace funnel, ads, pricing/discount and returns data to find winning and underperforming styles, diagnose causes, and recommend whether to replenish, push, test or exit.

Winning Styles Data Bank — link product/design attributes and images to market performance.

AI Style Search & Comparison — compare new styles to similar historical ones using attributes and images (multimodal), using existing models where they fit.

Style Selection & Early Winner Identification — rules and ML models for launch decisions, early winner signals and replenishment.

Business & Manufacturing AI — planning, quality, productivity and manufacturing-ops decision tools and workflow automation.

What you need

5–6 years in hands-on Applied AI, Data Science or ML Engineering roles.

Strong Python and SQL; works independently on large, messy datasets.

Solid statistics, applied ML and model evaluation.

Has used GenAI/LLMs to build real solutions, not just notebook prototypes.

Has shipped a usable tool or product (APIs, databases, lightweight apps such as Streamlit, FastAPI or a simple frontend).

Has personally built and deployed solutions — managing teams or producing analysis only does not count.

Comfortable working directly with business users in an ambiguous, fast-moving setting.

Nice to have

Retail, fashion, D2C, e-commerce, marketplace, supply chain or manufacturing experience.

Recommendation systems, image or multimodal AI (CLIP or embeddings, visual similarity search), agents.

Demand forecasting, assortment or merchandising analytics, pricing or markdown models.

Education

Tier 1 preferred. Tier 2 colleges are fine if the projects and experience are strong and varied.

Working model

On-site in Ahmedabad. Candidates willing to relocate are welcome.

Responsibilities

Own data and AI solutions end to end: problem, data, approach, build, test and iterate with users

Combine data from multiple sources to find patterns, diagnose performance and support decisions

Pick data science, ML or GenAI to fit the problem

Build usable tools — automated analysis, decision tools, AI workflows, simple apps — connected to databases and APIs

Turn one-time analyses into repeatable products and measure the value they create

Work directly with business and manufacturing teams