Software Engineer
Current- Designed AI system architecture handling document retrieval and real-time geospatial data integration.
- Optimized multi-stage inference pipelines, improving individual task accuracy by up to 24%.
- Developed an LLM evaluation framework measuring response quality, latency, and cost optimization across multiple model deployments.
- Automated deployment pipelines for managing and provisioning LLMs in secure environments, including air-gapped GPU servers and AWS GovCloud.
Feb 2023 - Present
