Thank you 😊

Dominik Kundel on LinkedIn: It's surreal to write this after almost 9 years, but today is my last day… | 22 comments It's surreal to write this after almost 9 years, but today is my last day at Twilio. It's been an unbelievable ride and I'm so grateful for all the… | 22 comments on LinkedIn

It's absolutely surreal but today is my last day at Twilio after almost 9 years and it's been a wild ride that I will forever be grateful for ❤️

www.linkedin.com/posts/dkunde...

Picture of the wait times at SFO with general queue being 8 minutes and TSA Precheck 10 minutes.

SFO is the only airport I know where Precheck has a longer line than regular security 😅

📌

Incredible photos 👏👏

Happy "unsubscribe from all the stores you didn't know you still consented to getting marketing emails from" day 😄

For example labelers like the Pronoun labeler or the GitHub Contributor labeler are fun:
bsky.app/profile/pron...

Bluesky is an excellent rabbit hole of potential 😂

I just use the pinned feed :) it's not great but works 😅

blueskydirectory.com/feeds/pins

Hi, so I've spent the past almost-decade studying research uses of public social media data, like e.g. ML researchers using content from Twitter, Reddit, and Mastodon.

Anyway, buckle up this is about to be a VERY long thread with lots of thoughts and links to papers. 🧵

I love the concept of building real apps with real users. Building apps like Twilio Barista was one of the best ways for me to learn all the edge cases of our products and also try products I hadn't tried yet.

Have you looked into custom feeds yet? They are great as well! I love the Quiet Posters one for example and I love that you can build your own

I love the house analogy!

💸 No. 10 - Unbounded Consumption

Unrestricted LLM usage can lead to denial-of-service attacks or excessive costs.

🔐 Mitigation tips: Implement rate limits, monitor resource usage, and throttle requests.

🤥 No. 9 - Misinformation

LLMs may propagate false or harmful content from biased or unverified sources.

🔐 Mitigation tips: Use fact-checking workflows and curate trusted data sources.

🔢 No. 8 - Vector and Embedding Weaknesses

Unsecured embeddings may expose models to poisoning or unauthorized access.

🔐 Mitigation tips: Encrypt embeddings, validate inputs, and restrict database access.

🔍 No. 7 - System Prompt Leakage

Exposure of system prompts can reveal application logic or sensitive information.

🔐 Mitigation tips: Avoid storing sensitive data in prompts; encrypt or obfuscate key instructions.

🪓 No. 6 - Excessive Agency

Granting too much autonomy to LLMs can enable harmful or unintended actions.

🔐 Mitigation tips: Limit permissions, add user oversight, and enforce action constraints.

🤝 No. 5 - Improper Output Handling

Unsanitized outputs can lead to XSS, SQL injection, or system-level attacks.

🔐 Mitigation tips: Sanitize outputs and enforce encoding based on context (HTML, SQL, etc.).

☠️ No. 4 - Data and Model Poisoning

Compromised datasets or tampered models lead to biased outputs or hidden backdoors.

🔐 Mitigation tips: Vet datasets, track transformations, and validate outputs against trusted sources.

⛓️ No. 3 - Supply Chain Risks

Third-party dependencies or tampered models can introduce vulnerabilities in LLMs.

🔐 Mitigation tips: Audit dependencies, enforce provenance checks, and validate model integrity.

🤐 No. 2 - Sensitive Information Disclosure

LLMs can leak private data or proprietary information via crafted queries or poor sanitization.

🔐 Mitigation tips: Mask sensitive data, restrict access, and monitor logs for leaks.

💉 No. 1 - Prompt Injection

Attackers manipulate LLM prompts to alter behavior, bypass security, or gain unauthorized control.

🔐 Mitigation tips: Use input validation, output constraints, and enforce principle of least privilege.

The new @owasp.org 2025 Top 10 Risk & Mitigations for LLMs and Gen AI Apps was released and there are some great updates in there! Highly recommend giving it a read.

If you are strapped for time, check the thread 🧵 for a short summary on each.

genai.owasp.org/llm-top-10/

Moin

You should turn them into a starter pack as well so it's easier for people to follow them 🙂

A bummer! I will see you at re:invent then I assume 🙂

Aw man I was going to come but ended up getting last minute concert tickets. Are you going to AI Tinkerers on Thursday?

How AI Agents Will Reshape Your Growth Marketing Strategy Twilio posts cloud communications trends, customer stories, and tips for building scalable voice and SMS applications with Twilio's APIs.

If you are looking for a written version instead check out my blog post on the Twilio blog about the same topic:

www.twilio.com/en-us/blog/a...

Twilio AI Assistants YouTube video by AI User Group

My talk from the "AI for Marketers" User Group just went live! In it I cover what an AI Agent is and how agents will change different aspects of Growth Marketing and Marketing at large.

youtu.be/2gU72m_kyAo?...