Performance-driven Machine Learning Engineer with over 5 years of experience building production AI systems and internal agent infrastructure. Proven ability to develop automated Large Language Model (LLM) workflows and agentic tools that enhance engineering productivity and streamline the Software Development Life Cycle (SDLC). Experienced in model selection, benchmarking, and operating end-to-end AI stacks to deliver high-impact results in fast-paced environments. Committed to supporting technical teams by implementing cutting-edge AI developer tools and instrumentation to measure velocity gains.
Performance-driven Machine Learning Engineer with over 5 years of experience building production AI systems and internal agent infrastructure. Proven ability to develop automated Large Language Model (LLM) workflows and agentic tools that enhance engineering productivity and streamline the Software Development Life Cycle (SDLC). Experienced in model selection, benchmarking, and operating end-to-end AI stacks to deliver high-impact results in fast-paced environments. Committed to supporting technical teams by implementing cutting-edge AI developer tools and instrumentation to measure velocity gains.
- Developed production-grade LLM workflows for internal transaction monitoring, utilizing Python and XGBoost to improve detection accuracy by 34%.
- Implemented real-time inference pipelines using Kafka and FastAPI to achieve sub-100ms latency for streaming data applications.
- Built internal infrastructure for evaluating model quality and performance of GPU-accelerated systems within high-throughput environments.
- Automated model monitoring and drift detection processes using Python and SQL to ensure high reliability across deployment cycles.
- Built agent-based applications for claim summaries using vector embeddings and Large Language Models (LLMs), reducing false positive alerts by 15% through iterative testing.
- Optimized database queries and data pipelines within Azure cloud environments, increasing overall processing efficiency by 20% for real-time applications.
- Developed custom internal tools to analyze conversational datasets and extract insights using Hugging Face transformers.
- Collaborated with cross-functional stakeholders to translate operational requirements into automated workflow processes for improved team productivity.
- Automated appointment reminders via Python scripts, saving 5+ hours of manual administrative work per week for clinic staff.
- Developed Python Flask REST APIs and Extract, Transform, Load (ETL) pipelines to manage and analyze healthcare datasets.
- Performed data-driven analysis on consultation datasets using SQL and Matplotlib to identify patterns for system improvements.
- Configured secure storage systems using Azure Blob Storage and implemented unit tests to maintain software quality standards.
• Developed a benchmarking framework to evaluate model configurations using vLLM, improving response accuracy by 25% across domain-specific queries. • Built an agentic AI assistant integrating OpenAI and Anthropic APIs with Pydantic for robust data validation and multi-agent orchestration. • Implemented session-based memory and dynamic prompt engineering to improve the reliability and context-awareness of internal system workflows.