cammasmith
cammasmith on Hacker News
Posted by
cammasmith on Wednesday September 09, 2026 @12:31PM
from the omnafy.com dept.
Really enjoyed reading your perspective. I am a recent math PhD graduate, and I also find it both exciting and terrifying to see what AI is doing to the field of math.
I used AI to help write the article, but the ideas and the technical perspective are mine. This wasn't made wholesale. I've been working on creating an MCP server for DynamoDB, and have been thinking a lot about how much better AI works with SQL than without it. Considering how many people use DynamoDB, it surprises me that there aren't better ways to get AI to interact with it. Just wondering if other people feel the same way.
Posted by
cammasmith on Thursday June 04, 2026 @11:06AM
from the ask-hn dept.
LLMs are good at SQL. It's precise, expressive, and unambiguous. If you connect an MCP server to Postgres, then the agent can query it directly. The same cannot be said for NoSQL, and given how many people use NoSQL databases, I’m surprised there isn’t more discussion about it.Part of the problem is diversity. MongoDB, DynamoDB, Cassandra, Redis, and Neo4j all have different query models....
Posted by
cammasmith on Thursday May 07, 2026 @09:57AM
from the show-hn dept.
Of all the NoSQL databases, DynamoDB is one of the most frustrating to query. MongoDB has an expressive query language. Cassandra has CQL, which is close enough to SQL that most engineers can pick it up quickly. DynamoDB gives you key condition expressions, attribute name placeholders for reserved words, and a query model where you have to pick the right index manually for every call. PartiQL...
Are there actually people out there who have never played “modern board games”? Maybe I’m just surrounded by nerds, but I don’t think these games are very niche. If someone I know hasn’t played board games like these, then it’s usually because they don’t have the attention span or don’t enjoy the strategy, not because they don’t realize board games exist.
I read this article assuming the title was clickbait, but no, that pretty much sums up the whole article. “Let’s solve the AI takeover using eugenics!” If even Elon musk and the Chinese government claim the solution is immoral, then all we really have is a second problem on our hands.
Posted by
cammasmith on Thursday April 23, 2026 @09:02AM
from the ask-hn dept.
My dad and I built a SQL layer for DynamoDB (DynamoSQL) because we kept running into the same problem: we needed to query information across multiple tables (or multiple record types within a single table) and had no good way to do it. We would either duplicate data across lots of records (denormalize everything upfront) and hope we guessed right, do multiple fetches and join in application...
Trade school seems to be looking more and more valuable these days. I feel bad for young people who are going to a 4 year university and gaining debt for a career that might not exist in 5 years. Being a plumber might not seem glamorous, but we're a long way off from AI learning to fix your toilet.
Interesting post. I'm notoriously bad at noticing the common characteristics in AI writing, but once they were pointed out, I realized I've been seeing them everywhere in websites.
Cool article. Did a very good job explaining things simply and providing good diagrams.
I have a hard time imagining that universal medicaid would ever become reality. A more realistic goal might be making the premium tax credit more accessible or lessening the restrictions to qualify. That way, more people could afford marketplace health insurance.
Can't even imagine what it's like to live in Europe. Just casually going on a walk and finding a coin that is over 2 millennia old. Just another Tuesday.
Maybe, but I'm not surprised by the number. Saying something triggering about a topic that's triggering from a man that's triggering is bound to get responses. It got me to respond to your response.
I think the price would need to be closer to $4 or $5 per letter for me personally to recommend it or consider it. My SIL just sent out wedding invitations with stamped seals, and I think she would have paid $5 per letter to mail 50 envelopes. But at $8 a letter, I start thinking about how I can pay someone to print the cards and I can buy a wax seal at Hobby Lobby myself.
Cool idea. I can see you guys doing well in the wedding industry. The price seems steep, but I appreciate the transparency of your pricing. It's refreshing.
That was a lot of fun. I got it wrong. I narrowed it down to two options, and neither was correct. Good stuff.
Posted by
cammasmith on Tuesday April 14, 2026 @08:54AM
from the show-hn dept.
https://dynamosql.com/
As a software developer, I've frequently used DynamoDB, and I have found myself missing the ease of SQL querying. So I built DynamoSQL to be a SQL layer. It allows you to write standard ANSI SQL in DynamoDB, including JOINs across tables, CTEs, subqueries, GROUP BY, and aggregations. There's no ETL, data export, or separate warehouse. The optimizer is index-aware and uses your partition keys,...
It kinda seems like the point of the article was to talk about different mathematical illustrations, not to determine if math was big or small. Even in the article, the conclusion is that it's both. I suspect the only reason for choosing the title is to grab attention (and it worked on me).
Of course, I am extra cynical as a number theorist who can't visualize most of my field. I wrote my doctorate on Siegel modular forms, and I can honestly say I have no way to visualize them any further than numbers on a page.
I agree for cases that need real-time replication and are querying at high frequency. There is a trade-off, though. You're adding three services to maintain, there's replication lag, and Aurora's running continuously whether you're querying or not. For teams that need occasional analytical access to their DynamoDB data without building and operating a replication pipeline, I still think the in-memory trade-off is worth it. Different teams, different trade-offs.
Valid points. The read cost and in-memory aggregation concerns are real and worth flagging. DynamoSQL is index-aware and will use your partition keys, sort keys, and GSIs where it can, but you're right that a well-designed native access pattern will be cheaper.
That said, not everyone using DynamoDB has the luxury of designing perfect access patterns from the start. Real production tables are often messier than the textbook examples. And for analytical or exploratory work (e.g. understanding your data, debugging a production issue, running ad-hoc reports) a slightly more expensive read is a reasonable trade-off for writing SQL in seconds rather than SDK code in minutes.
I appreciate the feedback. The cost visibility question is something worth being clearer about in the docs.
I agree. If Word had something like this, it would be so much easier to find the symbol you're looking for.
One of my biggest issues with playing DND is that I never fully understood the rules. I'd play with people who had been playing for years, and they didn't explain things very well, and that made it hard to play. Hopefully, this will help with that.
Built a SQL interface for DynamoDB. I was tired of constantly trying to find workarounds for querying on a NoSQL database, so I built a direct interface for DynamoDB, which I'm calling DynamoSQL. With it, you can use standard SQL (even JOINs) on DynamoDB. I'm really excited about it, and I'm starting beta trials this week.
https://dynamosql.com
Cool game. The sound keeps glitching on and off. Also, maybe I'm just dumb, but I couldn't figure out where to go or how to land.