Data Scientist

About Moative

Moative, an Applied AI Services company, designs AI roadmaps, builds co-pilots and predictive AI solutions for companies in energy, utilities, packaging, commerce, and other primary industries. Through Moative Labs, we aspire to build micro-products and launch AI startups in vertical markets. 

We have built and sold two companies, one of which was an AI company. Our founders and leaders are Math PhDs, Ivy League University Alumni, Ex-Googlers, and successful entrepreneurs.

Work you’ll do

As a data scientist, you will lead data-driven projects, design and develop advanced analytical frameworks and AI/ML solutions to address business problems. You will collaborate with product managers, engineers and domain experts to deliver intelligent solutions and products.

You’ll analyze new opportunities and ideas, evaluate new AI/ ML models/ frameworks/ platforms, conduct experiments, develop PoCs and prototypes.

As a Data Scientist, you will provide your advanced expertise on statistical and mathematical concepts and guide the team in AI/ML algorithms and model development. You will stay up-to-date with the latest advancements in data science, machine learning, and AI.

The ideal candidate will have a strong background in statistics, machine learning, and programming, as well as excellent business understanding and product design thinking skills. If you are passionate about data and have a proven track record of delivering impactful data solutions, we would love to hear from you.

Responsibilities

• Frame problems before you model them. You will define the problem structure, identify the right success metric, and map failure modes — data drift, integration cost, adoption friction — before a single model is trained. Post-mortems are not your primary output; pre-mortems are.
• Own delivery end-to-end, including deployment. You will take models from scoping through production. If your best work is a notebook that never shipped, this role is not for you. You will own the last mile: deployment, monitoring, iteration in production.
• Sit embedded in client teams and hold the room. You will join client standups, present modelling choices to client leadership, and defend or revise your approach on the spot. You will be the technical voice accountable for outcomes — not a back-office supplier of models.
• Build accelerators and reusable frameworks, not one-offs. You will identify repeatable patterns across engagements and convert them into tools, templates, and internal infrastructure that make the next delivery faster and more defensible.
• Write and communicate with precision across audiences. You will produce decision memos, model cards, and post-mortems that are specific enough for an engineer and clear enough for a CFO. You will cover trade-offs, assumptions, and risks — in the same meeting, for both rooms, without dumbing either one down.
• Drive ML lifecycle discipline. You will establish and enforce best practices across model development, versioning, evaluation, and monitoring — and raise the bar for how the team thinks about model quality and production readiness.

Who you are

You are a data scientist who is passionate about using AI/ML to improve processes, products and delight customers. You have experience working with less than clean data, developing ML models, and orchestrating the deployment of them to production. You thrive on taking initiatives, are very comfortable with ambiguity and can passionately defend your decisions.

Requirements and skills

• 2+ years of hands-on data science with shipped production models. Evidence of models that moved from development to deployment with measurable business impact. "Worked on" does not qualify — you must have owned the outcome.
• Consumer-scale domain depth in a regulated or operationally sensitive business. Direct experience with one or more of: credit risk and portfolio modelling (PD/LGD/EAD, scorecards, alternative-data underwriting, collections or behavioural scoring) or retail and commerce modelling (demand forecasting with seasonality and promo effects, assortment and markdown optimisation, customer segmentation and LTV, returns prediction, pricing elasticity). You can read a delinquency curve, or a sell-through curve, or a cohort retention plot and know what it implies for the next model decision and the next business decision
• Production GenAI and agentic system experience beyond prompt engineering. Hands-on with retrieval design, eval harnesses, guardrails, and fine-tuning vs. prompting trade-offs. You understand the observability and cost discipline required to run these systems in production. Prompt engineering alone does not qualify.
• Cloud and MLOps fluency. Proficient across at least one major cloud (AWS, Azure, or GCP) and experienced with MLOps tooling — MLflow, model registries, CI/CD for ML, and drift monitoring in production.
• Client-facing delivery experience. Has worked directly with external clients or business stakeholders — not just internal teams. Comfortable presenting technical choices, fielding pushback, and adjusting in real time without losing the thread.
• Structural problem framing, not just modelling skill. Demonstrates the ability to define what problem is actually worth solving, choose the right analytical approach for the business context, and articulate why alternative approaches were rejected.

Working at Moative

Moative is a young company, but we believe strongly in thinking long-term, while acting with urgency. Our ethos is rooted in innovation, efficiency and high-quality outcomes. We believe the future of work is AI-augmented and boundary less.

Here are some of our guiding principles:

• Think in decades. Act in hours. As an independent company, our moat is time. While our decisions are for the long-term horizon, our execution will be fast – measured in hours and days, not weeks and months.
• Own the canvas. Throw yourself in to build, fix or improve – anything that isn’t done right, irrespective of who did it. Be selfish about improving across the organization – because once the rot sets in, we waste years in surgery and recovery.
• Use data or don’t use data. Use data where you ought to but not as a ‘cover-my-back’ political tool. Be capable of making decisions with partial or limited data. Get better at intuition and pattern-matching. Whichever way you go, be mostly right about it.
• Avoid work about work. Process creeps on purpose, unless we constantly question it. We are deliberate about committing to rituals that take time away from the actual work. We truly believe that a meeting that could be an email, should be an email and you don’t need a person with the highest title to say that out loud.
• High revenue per person. We work backwards from this metric. Our default is to automate instead of hiring. We multi-skill our people to own more outcomes than hiring someone who has less to do. We don’t like squatting and hoarding that comes in the form of hiring for growth. High revenue per person comes from high quality work from everyone. We demand it.

If this role and our work is of interest to you, please apply here. We encourage you to apply even if you believe you do not meet all the requirements listed above.

The position is based out of Chennai. Our work currently involves significant in-person collaboration and we expect you to be present in the city