This takes me back, fighting telephone fraud back when folks use to accept cc over the phone. We used similar patterns but only had phone numbers and the white pages. Cross state boundaries inside similar time frames and categorizing similar merchant types. It’s fun to see these same patterns still in use 20 years later for the same purpose.
In the US I tend to agree (given the current pay to live system is constructed) but in the UK with single payer insurance this seems more palatable.
I’m curious if a “free society / libertarian” middle ground would be limiting access to NHS for those that choose to continue to use known harmful substances. I’d posit that many would object to that the way “death panels” were politicized when the Affordable Healthcare Act was passed though.
In my experience this isn’t the case. SaaS systems, at the least the ones that are embracing this sea change seem to recognize that connectivity is key. They are opening APIs and partnering like we’ve never seen. They lack the domain expertise and resources to plumb the last mile. Companies can finally customize and integrate there core SORs at a reasonable price point and if you hire the right people decent technology. It’s a golden era, I do wonder if you’re correct medium/long term but the lack of skill sets to build, deploy, and maintain these solutions is very real. Many can vibe code a great looking app, very few can support it day two even for just a few hundred users.
Lowering the bar to integrate and communicate is what has historically allowed technology to reach critical mass and enabled adoption. MCP is an evolution in that respect and shouldn’t be disregarded.
We had a non technical team member write an agent to clean up a file share. There are hundreds of programming languages, libraries, and apis that enabled that before MCP but now people don’t even have to think about it. Is it performant no, is it the “best” implementation absolutely not. Did it create enormous value in a novel way that was not possible with the resources, time, technology we had before 100%. And that’s the point.
We are in early stages of rolling out these capabilities to our team but we have aligned in VS Code + Cline + Sonnet 3.7. We use databricks but there’s a plugin for snowflake as well in vs code.
Couple of helpful tidbits we’ve learned:
- Define a read me that lays out your architecture. Normalization, libraries to use for different use cases, logging requirements, etc. reference it on every new task.
- keep notebooks from getting too big. A notebook per elt action or table definition make’s your compares faster and required context smaller.
Be prepared to be amazed. We’ve pointed Cline at an api doc and had it spit out elt code and a model that were nearly production ready. We’re pulling weeks off estimates…
I’d read that…I’d love to hear an Acquired podcast on the path of Long Distance Discount Services -> WorldCom -> MCI -> Verizon. They laid down a lot of fiber.
Curious how others feel about this narrative. As an SI we see the pain isolated systems cost businesses and the great deal of money required to make them fit into a larger ecosystem. That said as a rippling customer, their breadth and lack of depth is constantly on display. We’ve had several challenges with there “secondary” services. It costs money to be good at everything, my feeling is this style of getting off the ground is just as much a capital constraint as it is a revenue/gtm one.
He’s got a good valuation going selling ideas and figuring out how to implement later though…
The comment on shortening resumes resonates deep. The ability to differentiate skill set is almost entirely distilled to a secondary search if you happen to spot the tiny 2003-2010 indicating more than a passing affair. Not that it’s confirmation of talent but when all you see is the same gpt fueled write ups regurgitating your job posting it is indeed harder to prioritize and suss out quality applicants. I see the newer generation of recruiters pushing for a 15 second resume review, so the problem is certainly on both sides but it certainly cultivates the continued downward spiral.
The clinicians themselves can be pretty dreadful. The uses being explored are to give coaching to them ahead of delivering news to help boost empathy and increase understanding. “It looks like you had a subdural hematoma cause a status epilepticus seizure” vs “that fall you took broke some blood vessels in your head which caused a longer seizure than we like to see”.