Senior Applied Machine Learning Engineer

<h3>Accountabilities</h3><li>
<strong>Production Model Optimization:</strong>Improve existing machine learning models for churn prediction, customer retention, account health, and segmentation using structured customer data and product usage signals.</li><li>
<strong>Prediction Quality and Evaluation:</strong>Measure model performance against actual renewal, churn, expansion, and adoption outcomes to ensure predictions remain accurate and relevant to business decisions.</li><li>
<strong>Statistical Analysis and Backtesting:</strong>Develop time-based backtesting methodologies and fixed test sets to evaluate models rigorously and prevent data leakage from future information.</li><li>
<strong>Prediction Error Investigation:</strong>Analyze incorrect predictions at the individual customer account level to identify problems with labels, features, source data, or underlying assumptions.</li><li>
<strong>Data Quality and Feature Engineering:</strong>Investigate inconsistent CRM records and product usage data, address labeling issues, identify data leakage, and improve feature quality to strengthen model performance.</li><li>
<strong>Production Implementation:</strong>Translate analytical findings and model improvements into production-ready changes within the existing Go-based product codebase, ensuring that work extends beyond experimentation and notebooks.</li><li>
<strong>Model Monitoring and Automation:</strong>Within the first three to six months, automate evaluation processes and implement monitoring mechanisms that detect declines in recall, label quality, and product usage signals before they affect customers.</li><li>
<strong>Initial Model Assessment:</strong>During the first 30 days, conduct recall analysis for every existing customer using fixed test sets and implement improvements identified through the analysis.</li><li>
<strong>New Predictive Capabilities:</strong>Build, test, and validate additional models for expansion opportunities, cross-selling, forecasting, and marketing applications.</li><li>
<strong>Business-Focused Communication:</strong>Explain model performance, prediction errors, statistical findings, and recommended improvements in clear, accessible language for customers and stakeholders without a machine learning background.</li><li>
<strong>Data Confidentiality:</strong>Protect sensitive customer information, internal queries, training datasets, and analytical results when communicating findings or discussing model performance.</li><li>
<strong>Cross-Functional Collaboration:</strong>Work closely with backend engineering colleagues, particularly the Principal Backend Engineer, to maintain clear ownership boundaries and ensure prediction quality aligns with the wider product architecture.</li><li>
<strong>Independent Problem-Solving:</strong>Take ambiguous machine learning challenges from initial investigation through hypothesis development, testing, implementation, and production validation.</li><li>
<strong>Continuous Improvement:</strong>Establish repeatable evaluation practices and continuously refine models, datasets, and monitoring systems based on observed performance and real-world outcomes.</li><h2>Requirements</h2><li>
<strong>Professional Experience:</strong>At least five years of applied machine learning experience, specifically working with structured business data and production systems.</li><li>
<strong>Academic Foundation:</strong>Strong mathematical and statistical fundamentals, ideally supported by a rigorous academic background. A degree from an IIT or a similarly strong academic program is preferred.</li><li>
<strong>Production ML Ownership:</strong>Demonstrated experience developing, deploying, and maintaining production models that directly inform business decisions, such as churn prediction, retention, propensity modeling, customer health scoring, or segmentation.</li><li>
<strong>Machine Learning Fundamentals:</strong>Strong understanding of classification methods, precision, recall, model evaluation, fixed test sets, and time-based validation techniques.</li><li>
<strong>Data Quality and Leakage Prevention:</strong>Proven ability to identify and resolve data leakage, incorrect labels, weak features, and inconsistencies that can undermine model reliability.</li><li>
<strong>Business Data Experience:</strong>Practical experience working with complex, imperfect datasets from CRM systems and product usage platforms.</li><li>
<strong>SQL Proficiency:</strong>Strong SQL skills, including the ability to query, analyze, and validate large datasets to investigate model behavior and support feature development.</li><li>
<strong>Software Engineering Skills:</strong>Ability to work with Go application code and integrate machine learning improvements into an existing production codebase. Python may be used for analysis, but the ability to deliver production-level changes is essential.</li><li>
<strong>Model Evaluation and Error Analysis:</strong>Experience investigating individual predictions, identifying root causes of errors, and translating findings into concrete model or data improvements.</li><li>
<strong>Analytical Problem-Solving:</strong>Ability to independently structure ambiguous problems, develop testable hypotheses, conduct rigorous experiments, and implement practical solutions.</li><li>
<strong>Product Engineering Mindset:</strong>Experience working within a product-based organization and understanding how machine learning systems support real customer workflows and business outcomes. Candidates must currently be working at a product-based organization.</li><li>
<strong>Communication Skills:</strong>Ability to explain technical findings, statistical concepts, and model limitations clearly to non-technical stakeholders while maintaining appropriate confidentiality.</li><li>
<strong>Collaboration:</strong>Strong interpersonal skills and the ability to work effectively with backend engineers and other cross-functional stakeholders.</li><li>
<strong>Additional Advantages:</strong>A public GitHub profile, relevant machine learning side projects, or a technical blog demonstrating practical expertise and initiative would be a strong plus.</li><h2>Benefits</h2><li>
<strong>Remote Work:</strong>Full-time remote opportunity for candidates based in India.</li><li>
<strong>Competitive Engagement Structure:</strong>Full-time contractor arrangement with a US-based organization, with no deductions made by the hiring company from the agreed compensation.</li><li>
<strong>High-Impact Responsibilities:</strong>Direct ownership of machine learning prediction quality and the opportunity to influence customer-facing product decisions.</li><li>
<strong>Production-Focused Machine Learning:</strong>Work on live models and real business outcomes rather than limiting your contribution to research, prototypes, or experimental notebooks.</li><li>
<strong>Advanced Analytical Challenges:</strong>Solve practical problems involving statistical evaluation, data leakage, model accuracy, customer behavior, and prediction reliability.</li><li>
<strong>AI and Product Innovation:</strong>Contribute to new predictive capabilities for customer expansion, cross-selling, forecasting, and marketing.</li><li>
<strong>Cross-Functional Collaboration:</strong>Partner closely with backend engineering to integrate model improvements into production systems and deliver measurable product value.</li><li>
<strong>Technical Autonomy:</strong>Independently investigate complex problems, design experiments, and drive improvements from analysis through implementation.</li><li>
<strong>Professional Growth:</strong>Strengthen your expertise in applied machine learning, statistical modeling, production engineering, and automated model monitoring.</li><li>
<strong>Transparent Interview Process:</strong>Participate in focused technical discussions covering applied machine learning and statistics, model evaluation and error analysis, and SQL and product engineering.</li>