
6 min read
How many people can share one local AI setup? My DeepSeek numbers
One chat on two DGX Sparks gets 84 tok/s. Eight at once share 155 tok/s. What that means when you plan local AI for a whole team, not just one person.
Notes on getting AI working in businesses: what is worth automating, when to build, when to keep data on your own hardware, and how to check your website is not leaking.

6 min read
One chat on two DGX Sparks gets 84 tok/s. Eight at once share 155 tok/s. What that means when you plan local AI for a whole team, not just one person.
6 min read
Domain-driven design, plus services and repositories, gives AI coding agents clear names, one home for business logic and an easy line to test against.

5 min read
Cloud AI prices and hardware costs are likely to rise. How I match models to tasks and move work to local AI so my business isn't caught out.
5 min read
My local DeepSeek setup got 6.3s replies instead of 26.9s with 78% fewer tokens on computation tasks, and no gain on standard benchmarks. Here's why.

5 min read
Smarter models don't kill configured agents. They cut token waste, let you update one role everywhere, and let project captains run the work.

5 min read
I gave my local DeepSeek setup a way to run code mid-response. Same 38 of 40 coding tests solved, 16% faster overall, with fewer requests and tokens.

6 min read
Our app ran about 9.7k CI workflows in October. Compute would cost around $9 on one host, but plan limits on job starts would push it past $500.

5 min read
For the same money, two smaller machines beat one big one for local AI: faster first token, double the compute, no single point of blocking, better resale.
5 min read
One forgotten app out of 20 was breached and its keys were used to email about 25,000 people. Here's why usage limits saved me, and what I changed.

5 min read
I tuned DeepSeek-V4.1-Flash to run on two NVIDIA DGX Sparks. Here are the benchmark results, and what local AI means for cost, privacy and control.