---
title: "SQL Meets Slack: Doubt is a feature, not a bug"
newsletter: "MLOps Community"
date: 2025-03-06
source: https://aaif.live/newsletters/mlopscommunity/2025-03-06-sql-meets-slack-doubt-is-a-feature-not-a-bug
---

# SQL Meets Slack: Doubt is a feature, not a bug

*Plus, AI agents get specialized, RAGs get agentic, and NVIDIA speeds up drug discovery.*

*MLOps Community — Agentic AI Foundation, 2025-03-06*

Spend 4 years on a Computer Science degree, 5 years on a PhD, and rack up $150k in student debt to flip burgers at McDonald's [https://go.mlops.community/bv2vp5].

## Talk to Your Data: The SQL Data Analyst

[https://go.mlops.community/MLConfess](https://go.mlops.community/MLConfess)

## The Agent Exchange: Practitioner Insights

A Swiss Army knife or a scalpel? It’s a case of generalization or specialization, which I get into with Steven, Chiara, and Dmitri on AI agents.

Many see these agents as a threat to customer relationships and revenue models. Some platforms might try to block agents entirely, but long-term, collaboration through APIs seems inevitable.

Currently, specialized agents still dominate. Generic ones struggle with multi-step tasks, while specialized agents are faster and more reliable in real-time interactions. A hybrid approach is emerging:

 * Generalist agents hit limits quickly, failing on longer workflows
 * Specialized agents perform well but don’t scale easily across platforms
 * Hybrid models balance both, allowing generalists to handle orchestration while specialists execute tasks efficiently

With that in mind, I’ll leave you with the image of a surgeon using the corkscrew on his scalpel during surgery before you click to listen.

Video [https://go.mlops.community/jdqa56] || Spotify [https://go.mlops.community/bpkxoe] || Apple [https://go.mlops.community/ijav5z]

## Job of the Week

[https://go.mlops.community/jptp6v](https://go.mlops.community/jptp6v)

## BEST OF SLACK

There’s too much gold in Slack to cram in here, so we’re breaking it out into its own thing.

Don’t miss out - sign up here [https://go.mlops.community/BestOfSlack]!

Those who know me know I’m all for drug discovery. That's why this blog is worth a read.

It covers how NVIDIA’s BioNeMo slashes drug discovery timelines by reducing search space complexity, cutting computation time, and optimizing protein binder design with GPU-accelerated AI by:

 * Scaling AI workloads - Integrating AlphaFold2 and RFdiffusion, leveraging multi-GPU parallelism to accelerate structure prediction and sequence generation
 * Optimized inference - Using DiffDock 2.0 and ProteinMPNN to refine molecular docking and sequence design, improving efficiency without sacrificing accuracy
 * End-to-end automation - GPU-accelerated screening and stability assessments (ESM-1nv, ESM-2) reducing reliance on slow wet-lab validation processes

Have a read, and keep experimenting with drugs.

Read it here [https://go.mlops.community/cpofot]

## HERE TO HELP

Before you go, here are three ways I can help - just hit reply:

 * Curated intros to other community members
 * What problems are you dealing with? Let me help you find the best solutions through my network
 * Looking to augment your staff for an MLOps or AI project? I got you covered

Thanks for reading, catch you next time!

Interested in partnering with us? Get in touch: partners@mlops.community

Thanks for reading. See you in Slack [https://go.mlops.community/slack], YouTube [https://www.youtube.com/channel/UCG6qpjVnBTTT8wLGBygANOQ?view_as=subscriber], and podcast [https://home.mlops.community/public/content/] land. Oh yeah, and we are also on X [https://twitter.com/mlopscommunity] and LinkedIn [https://go.mlops.community/linkedin].

The MLOps Community newsletter is edited by Jessica Rudd [https://www.linkedin.com/in/jmrudd/].

---
Source: https://aaif.live/newsletters/mlopscommunity/2025-03-06-sql-meets-slack-doubt-is-a-feature-not-a-bug