@greenpants, "Pi coding harness but containerized and sandboxed" care to address some specifics and/or reference implementation for this. may be a GITHub URL?
If a start-up stage company is just getting started on their AI/ML journey has budget for 3 FTE people. The company already has traditional ETL/BI expertise and a "DWH". Who would you hire (data scientist, ML engineer, Dat Engineer) and how would you allocate the division of responsibilities?
1. Depending on your application, the end users can be from different target market/background. If that will be the case with your app, list down the top X markets and create a specific landing page which "talks" the language of the target market.
2. Cold-outreach, Find your ideal target customers on LinkedIn/Twitter, google them, message/email them on social media (lead finding tools) and ask for help. Be willing to offer to pay them for 10-15 minutes of their time. At least a few will help without asking for money.
3. Assuming what you are selling is described on a landing page (doesn't have to be), you can do a user test by asking questions to consumers using survey tools)
the goal is find out if users understand what you are trying to sell (clarity of message, trustworthiness)
You can use tools like Survey Monkey, Google Surveys, even Facebook ads.
here are a couple of examples of a purpose-built tool for feedback called ninjafeedback:
@legg0myegg0 thanks for bringing up Dremio. Can Dremio connect with different databases RDBMS, Kafka, DataLake PRC file formats? Is the use case limited to certain data stores? Is the use-case primarily a united query engine (so that the code remains consistent across DB engines) or is the use case query acceleration?
@gwittel, appreciate you sharing your insights. Will you be able to elaborate on "RDBMS will have natural limitations"? Can you provide a specific example?
Thanks for your insights. Great points on inertia and lack of big sales budgets. Agreed also on HDFC/data lake use cases with PQ files. However, regarding querying RDBMS, are you saying that Presto requires in ODBC/JDBC connectivity? Does Presto have an ability to connect with "native DB" drivers?
Serverless is great or it sucks are extreme viewpoints. Reality is somewhere in the middle. For some specific use cases, (image manipulation, Alexa skills, etc.) serverless is fantastic. Dealing with persistent storage and DBs is challenging. But continues to evolve.
Managing server fleets and keeping them patched is no small feat. How much of the productivity can we gain if we didn't have to do all that with serverless? It's not possible for all use cases today. But can we move from say less than 5% of the use cases today to say, 40-50% of the use cases in another 5 years?
if you are doing customer reporting, sooner or later you will need these two priority things ( more so than charting features)
1. Delegated authentication ( think Okta)
2. Fine-grained row-level security to slice the same data in the report based on authorization.
We have had good experience with Tableau but had to do some custom Rest API programming for #1 .. Feel free to PM