I'm Jesika, CEO of LogicLoop - an AI SQL Copilot. Ask me anything!
Jesika Haria
11 replies
Hey PH Community! š
I'm Jesika, a data veteran š©āš» I've been in this industry for over a decade - worked on Graph Search at Facebook, built Google Cloud's DataPrep at Trifacta, built OCR & data extraction tools for Top 10 Banks at Instabase.
But these last few months have been the most exciting ones of my career yet - we're working on an AI SQL Copilot to help business and data teams generate, fix and optimize SQL directly on their data schema.
And the best part? You can try it for yourself on May 23rd when we officially launch on Product Hunt!
But, before then, let's chat about how you can empower anyone to ask data questions at your company, LLMs in SQL workflows, and taking actions on your data.
I am here to answer all of your questions. So fire away! š
Replies
Jacqueline Xu@jacqueline_xu
How much do you see these newer developments in AI augmenting data and business users' jobs vs replacing them?
Share
@jacqueline_xu Good question - a lot of analysts are stuck being "data monkeys", aka just responsible for pulling data without much analysis, and we help them to do higher-leverage work with their data instead of just pulling ad-hoc queries for business users or exec reports. I don't see data roles going away anytime soon!
@dllp Right, here's some ways we're different -
š On your data schema: A lot of the ones we've seen recently don't actually generate these queries on your schema, so you still have to do a lot of manipulation or calculation to get to SQL that just produces results.
š» A better experience in productionizing SQL: Like code, writing production SQL is rarely one-shot. LogicLoop comes with a suite of AI SQL Copilot tools, like a fixer, optimizer, editor etc. that allow you to continue to iterate on your SQL logic in an integrated environment.
šØ Go beyond SQL: Generating query results might often be the first step of a business process. With LogicLoop, you can actually visualize results, send Slack alerts, create tickets, run automations etc. on a schedule.
You'll be able to see all this live soon! š
How does it accommodate users with varying levels of technical expertise? Can non-technical users effectively utilize the tool?
@ben_levine13 Users that generally understand where their data is but are not SQL-confident can easily use this tool now! If you don't have data access at your company, most tools aren't going to be able to help today. (But we're hoping seeing early success with tools like this will change that culture!)
How accurate have you seen the tool be?
@jesika_haria How effective is the query quality in the tool? During my testing, I noticed occasional instances where it selected incorrect columns or tables. However, upon reiterating the instructions, the tool successfully generated the appropriate query.
You mentioned the availability of AI SQL Copilot tools, including fixer and optimizer. Could you please elaborate on these tools? Are they external third-party AI tools, or are they proprietary tools developed by your team?
Additionally, I'm curious to know if you are utilizing the OpenAI completion (API-based) model or a finetuned model for these capabilities?