• Own and develop the product analytics function • Work closely with Product, Engineering, and Leadership teams • Proactively identify what should be analyzed, validated, or challenged — without waiting for requests from Product or Leadership • Run deep product analysis to identify opportunities for growth and optimization • Drive product decisions with data, not just respond to ad-hoc requests • Design event tracking from scratch and maintain scalable, reliable analytics instrumentation • Own data quality and metric consistency across the product • Improve and optimize the DWH layer, including data modeling and performance optimization • Build and maintain dashboards and analytical layers • Develop product metrics, frameworks, and measurement approaches • Help teams make faster and better product decisions through data

• Design and maintain the core AI orchestration layers, centralized API gateways, and reusable frameworks (e.g., advanced RAG architectures, agentic frameworks) for company-wide consumption. • Collaborate directly with core engineering teams to embed production-ready generative AI and machine learning features into the Actian Data Intelligence Platform. • Establish strict guardrails, evaluation frameworks, and monitoring tools to track model performance, bias, data privacy, and security across all AI implementations. • Actively monitor and manage cloud and API compute spend (token management, open-source vs. commercial models) and optimize execution latency for production AI features. • Lead workshops, design blueprints, and create documentation to empower non-AI engineering teams to build and maintain their own AI-driven features confidently. • Drive the engineering execution of high-impact AI proof-of-concepts, ensuring they are built with production-grade code that scales seamlessly. • Define and enforce the organization's official AI stack, from vector database selection and vector embeddings strategies to semantic caching mechanisms. • Evaluate and manage partnerships with AI model providers and lead the technical assessment of cutting-edge open-source models to keep Actian at the vanguard of innovation. • Define and track operational metrics for the AI Enablement function, such as developer adoption rates, reduction in time-to-market for AI features, and ROI of implemented AI tools.

• Design causal inference and quasi-experimental approaches for pricing and conversion measurement across our products • Build and calibrate probability and conversion models that feed optimization routines across our marketplace, distribution, and pricing products • Build elasticity and gain/loss models to power bidding and pricing optimization decisions • Select and validate statistical methods that best fit the data and business problem at hand • Build ranking and relevance models for sort order and personalization across our flight and hotel search products • Partner with engineering to productionize models, and with analytics and product teams to translate business questions into measurable, testable hypotheses

Role Description We are seeking a Data Scientist II (DS-2) to join our core data science team. In this role, you will build and ship models that power Mrsool's quick-commerce marketplace, owning well-scoped problems end-to-end — from analysis and experimentation through to production. You will work closely with cross-functional teams and senior data scientists to deliver robust, data-driven solutions. This position offers an opportunity to grow your craft on high-impact problems and contribute directly to the growth and success of the organization. What You Will Do - Marketplace Modelling: Build and maintain ML and optimisation models across the quick-commerce stack — supply-demand matching, dynamic and surge pricing, recommendations, ETA prediction, and broader marketplace optimisation. - Butler & Conversational AI: Contribute to the AI behind Butler, Mrsool's distinctive conversational ordering experience — modelling customer intent from free-form, unstructured requests (text, voice, images) and mapping it to fulfillable, well-priced orders. - Experimentation & Causal Inference: Design and run experiments (A/B and quasi-experimental) across pricing, matching, recommendations, and Butler, and turn noisy marketplace data into decisions stakeholders can act on. - Feature Engineering & Data Craft: Engineer high-signal features from messy, real-world data — order events, courier traces, geospatial signals, pricing configs, and conversational text/voice — as a core, ongoing part of the role. - Production ML: Own your models through their lifecycle — data pipelines, training, deployment, monitoring, and retraining — and respond when a model or config drifts. - Cross-Functional Collaboration: Collaborate effectively with product managers, engineers, DevOps, operations, and other squads to deliver seamless, data-driven experiences and to help diagnose live issues (e.g. mispriced brackets, elevated failure rates in a city). - Operational Excellence: Proactively monitor model and metric health, instrument your work with proper logging and observability, and contribute to reliable, repeatable analysis and deployment practices. - Continuous Improvement: Identify opportunities to improve measurement, modelling, and process; favour small, incremental changes that compound over time. Qualifications - Years of Experience: 3 to 4 years of non-internship professional data science or ML experience in fast-paced product startups or high-scale tech enterprises. - Experimentation & Causal Inference: Solid command of A/B test design, power analysis, and quasi-experimental methods (diff-in-diff, instrumental variables, synthetic control), including awareness of interference in marketplace/network settings. - ML & Optimisation Depth: Strong grounding in forecasting and at least one of operations research / reinforcement learning applied to allocation, matching, or pricing problems. - Feature Engineering: Proven ability to build, select, and maintain features from large, messy, real-world data. - Production Engineering: Comfortable deploying, monitoring, and maintaining ML pipelines, with the engineering discipline to keep models reliable in production. - Technical Toolkit: Fluent in Python and SQL, with the ability to work efficiently against large-scale data. - Problem-Solving Mindset: A knack for thinking from first principles and a track record of delivering high-quality work while balancing trade-offs like reliability, latency, and interpretability. - Iterative Mindset: A bias towards shipping early and iterating; a belief in small, incremental changes over large, multi-quarter undertakings. - Education: Bachelor's/Master's degree in Computer Science, Statistics, Engineering, or an equivalent quantitative field. Who Will Excel - Data scientists with hands-on experience in quick commerce, marketplaces, logistics, ride-hailing, or on-demand delivery, who understand two-sided supply/demand dynamics. - Those with NLP / LLM experience — intent classification, entity extraction, embeddings, or conversational/voice data — directly relevant to Butler. - Engineers comfortable with streaming/big-data tooling (Spark, Kafka) and real-time inference. - High-agency individuals who treat their models as products and collaborate well across conflicting perspectives. Benefits - Inclusive and Diverse Environment: We foster an inclusive and diverse workplace that values innovation and offers remote environments. - Competitive Compensation: Our compensation packages are highly competitive and include potential share options for certain roles. - Personal Growth and Development: We are committed to your personal and professional growth, providing regular training and an annual learning stipend to help you advance your career in a dynamic environment. - Autonomy and Mentorship: You'll enjoy a high degree of autonomy in your role, supported by mentorship and ambitious goals that pave the way for both your success and the company's growth.