Markdown source

Dense Slides: Trust, but Verify: Shreya Rajpal

Source Video

Trust, but Verify: Shreya Rajpal

Method

This deck is slide-only. The existing captured video frame set supplies candidate frames, then local OpenCV rejects sponsor/title/speaker-only frames, crops visible slide surfaces, deduplicates, and saves the cropped slide images.

Cropped Visible Slides

slide-001.jpg

Slide text:

About me

Current CEO & Cofounder @ Guardrails AI

Past ML Infra lead @ MLOps Co,

ML @ Self driving cars,

Classical AI & Deep learning research

slide-002.jpg

Slide text:

We're seeing a cambrian explosion of applications in AI

Auto-GPT May Be The Strong AI Tool That Surpasses ChatGPT

Seriously, software engineering as we have known it is dead in the water.

This AI is better than at least 95% of coders. And to the extent it is not, it delivers 1000x productivity improvements to complement them.

We have now officially entered the Age of AI.

CAN AI TREAT MENTAL ILLNESS?

How Generative AI Will Change Sales

slide-003.jpg

Slide text:

The reality

Path to 100 Million Users

One Month Retention

Incumbents

AI-First Companies

Source: https://www.sequoiacap.com/article/generative-ai-act-two/

slide-004.jpg

Slide text:

The reality

Path to 100 Million Users (stylized)

-2 months 9 months 30 months 41 months 55 months 61 months

Months from launch

One Month Retention

Incumbents

AI-First Companies

63% Median

42% Median

Guardrails AI

Source: https://www.sequoiacap.com/article/generative-ai-act-two/

slide-005.jpg

Slide text:

Alex Graveley

@alexgraveley

Simple LLM technique that helps a lot (but you might not be using): add a constraint checker to ensure valid generation. On violation, inject what was generated and the rule violation, and regenerate.

slide-006.jpg

Slide text:

Guardrails AI acts as a safety firewall around your LLMs

slide-007.jpg

Slide text:

Guardrails AI under-the-hood

Creating a guard

Select type of output to validate

RAIL Spec

Pydantic Model

String

LLM Callable

Add prompt & instructions

Initialize guard from spec

Invoke Guard

Calling a guard

Guard invokes LLM API

LLM API Returns

LLM Output is validated

Invalid

Valid

Logs

Return output

Guardrails AI

slide-008.jpg

Slide text:

What Guardrails AI does

Guardrails AI is a fully open source library that offers

Framework for creating custom validators

Orchestration of prompting → verification → re-prompting

Library of commonly used validators for multiple use cases

Specification language for communicating requirements to LLM

slide-009.jpg

Slide text:

What Guardrails AI does

Guardrails AI is a fully open source library that offers

Framework for creating custom validators

Orchestration of prompting → verification → re-prompting

Library of commonly used validators for multiple use cases

Specification language for communicating requirements to LLM

slide-010.jpg

Slide text:

How do I prevent LLM hallucinations?

Provenance Guardrails

Every LLM utterance should have a source of truth.

slide-011.jpg

Slide text:

Example: Validating "correctness"

Application Logic

Guardrails AI

LLM Logic

Prompt

LLM API

Raw Output

Reconstruct Prompt

Fail Validation

Verification Logic

Pass Validation

How do I change my password?

Raw Output

1. Log into your account.

2. Go to user settings by clicking on the top left corner.

3. Click change password.

slide-012.jpg

Slide text:

More examples of validations

Make sure my code is executable

Never give financial or healthcare advice

Don’t ask private questions

Don’t mention competitors

Ensure each sentence is from a verified source and is accurate

No profanity is mentioned in text

Prompt injection protection

Never expose prompt or sources

slide-013.jpg

Slide text:

Learn more

Github: github.com/ShreyaR/guardrails

Website: guardrailsai.com

Twitter: @ShreyaR or @guardrails_ai

Classification audit: raw/sources/slide-ai-classification/dense/9-vGxMoUM9Y/audit.json