Kenneth Holstad

kenneth-holstad

·

kenneth-holstad

AI & ML interests

LLM evaluation, RAG, agents, ML for data science

Recent Activity

updated a Space 5 days ago

kenneth-holstad/lab11-send-it-to-fellowship

published a Space 5 days ago

kenneth-holstad/lab11-send-it-to-fellowship

updated a dataset 5 days ago

kenneth-holstad/lab11-send-it-to-fellowship

View all activity

Organizations

kenneth-holstad 's Spaces 12

Send It to Fellowship: a LoRA fine-tune, scored on voice and on substance

Compare before‑and‑after fine‑tuned nurse advice texts

Make It Say No: three validation layers, measured in both directions

Answer pharmacy questions with safety checks

Linear Regression from Scratch: Gradient Descent vs OLS

Compare gradient descent and OLS linear regression results

Build the Paperwork Machine: a ReAct agent, traced step by step

Answer medical queries using a step‑by‑step reasoning agent

Answer the Patient: what a reranker changes, measured

Answer medical questions using a reranked document search

Catch the AI Lying: an audited pass/fail harness for one small free model

Review AI model answers and see pass/fail results

Order the Lab: a tool-calling trace, and what the model said with no tools

Compare AI drug queries with and without tool lookup

Where the Money Goes: healthcare spend concentration and SDOH

View key healthcare spend metrics instantly

Anonymous, But How Safe? A k-anonymity risk meter for patient records

Explore de‑identified patient records and risk

What is something someone new to healthcare data might be surprised to learn?

Generate answers and factor charts for your questions

GP versus the Specialist: a measured model-tier scorecard

Compare AI model performance on clinical questions

Speak the Patient's Language: semantic search vs keyword search

Search clinical sentences using semantic embeddings