LLMs for Parents
Ck . Ck .

LLMs for Parents

I recently used ChatGPT-4o to help out with my 10-year old's math homework. I know there is much coming on the LLMs for education front but here's how you can use it it today:

Read More
Which is the best LLM
Ck . Ck .

Which is the best LLM

Which is the best LLM? Clearly a loaded topic with lots of strong opinions. The best LLM is probably one you've never heard of...yet... but brings all the pieces together. Let me explain…

Read More
Spellcheck for social interactions
Ck . Ck .

Spellcheck for social interactions

As GenAI becomes more integrated in our work lives, it's interesting to reflect on how it affects our personal interactions. For me, Microsoft Co-Pilot has been a game changer. It's a built-in tool that I use heavily, especially when summarizing large documents or compiling information from multiple sources. But what's most surprising is how it's transformed the way I respond to emails.

Read More
LLM for Healthcare Administration
Ck . Ck .

LLM for Healthcare Administration

In recent years, the healthcare sector has experienced a notable discrepancy in the growth rate of physicians compared to administrators. This expanding administrative workforce, although crucial for operational management, has contributed to ballooning healthcare costs. The potential for cost-cutting through administrative reduction is significant, and one promising avenue to achieve this is the integration of Artificial Intelligence (AI) in automating numerous administrative tasks.

Read More
The quantized (local) future
Ck . Ck .

The quantized (local) future

Over the last few weeks, I've been playing with quantized versions of LLMs and exploring options to run LLMs locally. I am using a MacBook Pro with 1 TB of storage and 16GB of RAM. Typically, that would not be enough to run a more powerful LLM locally but it's very manageable if you run a quantized model. Quantization in the context of Large Language Models (LLMs) like ChatGPT refers to a technique used to reduce the precision of the numbers (weights) in the neural network model. This is done to decrease the size of the model and increase its operational efficiency, especially in deployment scenarios.

Read More
The Economics of Software Development for Global Health
Ck . Ck .

The Economics of Software Development for Global Health

Much of the digital practice in Global Health is dominated by single use, bespoke software solutions. These are often branded and marketed as ‘platforms’. The proliferation of these ‘solutions’ has led to the creation of single use software packages that do not scale outside of a single geography and/or disease area. When they do, they need to be completed rebuilt from the ground up. Technologies that run on these solutions then need to be reconfigured costing considerable amounts and requiring considerable time and expertise.

Read More