Beyond the Demo
11 posts found
I always assumed higher temperature equals more hallucination in LLMs. Seems obvious. But I’ve learned that running on assumptions in data science is a fast way…
Four gaps between a promising prototype and a system people can rely on.
Useful clinical AI starts with clear boundaries, reviewable evidence, and accountable decisions.
Make the evidence behind generated research summaries easy to locate and verify.
An illustrative batch workflow with validation, review queues, and clear publishing boundaries.
Layered checks make missing evidence and summary errors easier to catch before review.
A sample design review of bounded tools, approval steps, and recoverable seller workflows.
Start with representative examples, explicit expectations, and a place to record failures.
Add separate agents only when the task needs distinct responsibilities and a useful handoff.
Separate reference existence, metadata accuracy, and support for the surrounding claim.
A release checklist for evidence, boundaries, operations, and ownership before a wider rollout.
A newsletter by Keerthi
Practical lessons on shipping AI people can trust. Evaluation, RAG, and agents across healthcare, publishing, and ecommerce.
New issue every two weeks · Unsubscribe anytime