Andrew Reed | Notion

Applied AI & Machine Learning Engineer

Contact

www.andrewreed.com

</aside>

@andrewrreed

</aside>

[email protected]

</aside>

linkedin.com/in/andrew-r-reed

</aside>

About Me

Hi, I’m Andrew, a machine learning engineer and technical leader with 10 years of experience designing, researching, developing, and deploying machine learning & AI solutions.

My expertise centers on architecting scalable ML systems, leading technical integrations with cloud partners, and translating research into practical applications. I've led multi-month research cycles, built enterprise-grade inference infrastructure, and created custom AI solutions that span fine-tuning, evaluation frameworks, and deployment automation.

I'm passionate about solving complex technical challenges and helping organizations realize the full potential of AI through thoughtful engineering and strategic implementation. I am inspired by simple ideas that are well executed.

Skills

Core

Experience

Hugging Face

Collaborating closely with customers to understand their unique challenges, design and implement custom ML + AI solutions, and ensure successful integration and optimization of production workloads.

Machine Learning Engineer, Cloud Integrations

Hugging Face ∙ Feb 2025 - Present

Technical lead for enterprise cloud partnerships, architecting production-ready ML infrastructure, and building integrations that enable organizations to deploy state-of-the-art AI systems at scale through cloud platforms.

Technical Lead for Dell Partnership

Primary technical point of contact for Dell Enterprise Hub (DEH) development activities

Led the design and development for Dell Enterprise Hub 2.0, architecting and implementing production LLM inference infrastructure for enterprise customers

Expanded model catalog support for 12 new models including priority Llama 4 integration, making Dell Enterprise Hub first partner with availability within hours of official release

Designed and built App Catalog integration featuring OpenWebUI + AnythingLLM with configurable MCP servers for enterprise deployment

Resolved 31 critical bugs with our LLM inference containers across multiple hardware backends (AMD MI300x, NVIDIA H200, Intel Gaudi 3) and inference frameworks (TGI, vLLM, SGLang)

Developed the dell-ai CLI/SDK and dell-helm-chart enabling automated enterprise deployments with Docker/Kubernetes

Actively contributed to the DEH portal Svelte app to accommodate changes and facilitate new feature releases

Technical Lead for Nutanix Partnership

Led the integration of Text Generation Inference (TGI) for Nutanix’s GPT-in-a-Box solution

Responsible for building, testing, benchmarking, validating, and optimizing model-specific inference server configurations for new model releases

Machine Learning Engineer, Customer Success

Hugging Face ∙ May 2023 - Feb 2025

Solution architect and technical advisor for enterprise AI implementations, working closely with startup to Fortune 500 customers to design, optimize, and deploy custom ML solutions.

Serviced $1.5M+ annual revenue across 16+ enterprise customers while achieving 8.88 average personal NPS score

Led pre-sales technical discussions resulting in $1M+ deal closure through comprehensive solution scoping, technical validation, and customer communication

Created 6 distinct workshop curricula (LLM Evaluation, LLM Inference, LLM Training/Fine-tuning, RLHF, RAG, Semantic Search) that were delivered 19 times to enterprise teams

Misc Open Source Contributions

Initiated and built the integration for the first LLM observability tools on the HF Hub: Arize AI’s Phoenix as HF Space & Langfuse as HF Space

Migrating from OpenAI to Open LLMs Using TGI’s Messages API + blog post

Built initial LangChain integration for HF chat models

Created 20+ example notebooks

Project Highlight: Performance Optimization for Patent & IP Startup

Designed fine-tuning approach for embedding models using synthetic and human curated patent data, achieving 23% accuracy improvement over baseline

Optimized a bulk embedding solution for 150M+ patent documents with Text Embeddings Inference (TEI) achieving 9x improvement of 2,700 embeddings/second throughput via multi-replica load balancing

Architected scalable semantic search solution to handle 30B+ embeddings with disk-based storage optimization

Project Highlight: Conversational Agent for SaaS Startup

Built conversational AI agent MVP using Anthropic API and smolagents library for prospect intelligence workflow that dynamically routes between automated company research (web scraping, data extraction) and interactive user interviews to generate tailored entity blueprints for sales demo personalization, deployed as Hugging Face Space with Langfuse observability integration

Fine-tuned custom NER model using SpanMarker, achieving 95% F1 score (9-point improvement) by addressing distribution gaps with augmented training data

Project Highlight: Process Automation with LLMs for Healthcare Startup

Architected LLM workflow to automate student psychological evaluation report generation and fact-checking pipeline to minimize hallucinations which reduced manual research & writing time from hours to minutes

Fine-tuned a vision encoder-decoder model (Donut) to transcribe handwritten essays while preserving spelling/grammar errors that improved character error rate (CER) by~4% over leading OCR solutions

Cloudera Fast Forward Labs

Responsible for staying up-to-date on latest trends in ML, identifying research topics based on advances in academia, and leading research projects to demonstrate how these advances can be applied in practice.