Senior/Lead AI/ML Engineer

Overview

We are seeking a Senior/Lead AI/ML Engineer to join our team in Hyderabad. The role requires a seasoned professional with extensive experience in machine learning, data processing, and applying advanced algorithms to real-world commercial datasets. A degree in Computer Science, Engineering, or a similar relevant field is essential, along with over 8 years of professional experience.

Key Responsibilities

• Design and implement a tiered matching system starting with deterministic rules, followed by string-distance and machine learning matching, and finally model-based classification for unresolved cases. Provide detailed reporting of coverage and accuracy at each tier and overall.
• Develop confidence scoring mechanisms and establish threshold levels that determine user interaction outcomes. Manage review queues to ensure that erroneous matches do not lead to incorrect pricing and handle client inquiries on pricing decisions.
• Implement and maintain an analyst feedback loop where override actions are utilized as signals to monitor and address quality degradation and model drift over time.
• Document and encode existing analyst matching rules, including exclusions from benchmarking, handling of low-volume items via broader pricing strategies, and price validation checks done by operators.
• Query and transform point-of-sale and supplier datasets within Compass Snowflake, operating primarily on the read-only source layer maintained by another team.
• Manage persistence of mapping rules, analyst overrides, and market basket versions in an operational store, planned through Aurora Serverless v2 with PostgreSQL awaiting final approval.
• Integrate machine learning models via Amazon Bedrock under established AI control policies within the Compass account, adhering strictly to compliance and governance commitments.
• Collaborate closely with pricing subject matter experts who guide and validate the matching decisions throughout development.

Required Skills and Experience

• Expertise in Python programming, particularly using libraries such as NumPy, Pandas, and Scikit-learn for data manipulation and machine learning.
• Hands-on experience in production environments focusing on text classification and entity matching. Familiarity with string-distance and similarity algorithms like Levenshtein, Jaro-Winkler, TF-IDF, cosine similarity, as well as vector embeddings applied to complex, real-world datasets.
• Proficient in advanced SQL for complex queries against cloud data warehouses, specifically Snowflake, including reading data managed by third-party teams.
• Strong background in deploying machine learning solutions with capabilities in drift detection, human feedback integration, and confidence monitoring across prediction results.
• Experience managing evolving taxonomies or master catalogs, supporting version control and data migrations in dynamic environments.

Preferred Additional Qualifications

• Experience with large language models used for classification or entity extraction on brief, noisy commercial texts with full auditability.
• Industry knowledge in retail, grocery, restaurant, or foodservice product data to accelerate understanding of product equivalencies and inventory descriptions.
• Familiarity with Amazon Bedrock or compliance-driven AI frameworks featuring model version pinning and comprehensive prompt and decision logging.