Matt Landers

If your MCP just wraps existing APIs, it’s time to rethink. MCPs should be built for LLMs—resolve ambiguity, handle name matching, and return what the AI really needs. Don’t force multi-step calls or vague errors. One intent, one call—build smarter, not harder. #MCP #LLM #AI #ProductDesign #DevEx

14

2 Comments

This is a great experiment by the Anthropic team, showcasing the "true autonomous" capabilities of LLMs. By "true autonomous," I mean that Anthropic (I believe) purposefully did not bombard the LLM with numerous tools in areas where LLMs are weak. For example, LLMs are not good in math, and performance would have been much better if they hd provided a tool like "calculate_profitability." This might have prevented the $100 opportunity miss. In AI agent development, tools bridge the gap where LLMs are not "deterministic." This is currently the MOAT in the GenAI app business: the better your tools, the more deterministic results you can provide. Only till LLMs can write their own tools on the fly and execute them where they know they are weak and using deterministic softwares (tools) to fill these gaps. https://lnkd.in/gMk3VRJv

10

1 Comment

A topic I will be watching closely in 2026 will be cost efficiency improvements for agents and AI. Over the fall, we saw a number of new efficiency features across the training, model, and agent layers. But in addition to new technologies from the likes of MongoDB (embedded models), Microsoft (AI Foundry), and Amazon Web Services (AWS) (Bedrock AgentCore), we should also expect some vendors to pass their own savings from efficiency gains back to customers. It looks like GitHub got the memo with this little holiday gift. #Github #agents #Microsoft https://lnkd.in/eiBBkxe2

2

Important thoughts on LLMs and Knowledge preservation. If we want aligned AI, we need to align it to grounded truths and curated human knowledge. That was the original promise of the web.

98

1 Comment

We open sourced our PII Classifier. The version of our PII model that we're open sourcing today has a 98% recall of potential PII conversations in English text on Roblox. The model has also achieved an F1 score of 94% on our production data, outperforming other state-of-the-art safety models, like LlamaGuard v3 8B (28%) and Piiranha NER (14%). Excited for the community to use it and hear feedback !

86

1 Comment

Insightful guide on building multi-agentic meta-tooling systems using AWS Strands SDK . It's fascinating to see how Strands simplifies the orchestration of specialized AI agents, leveraging models like Claude 4 Sonnet and Opus to tackle complex financial analyses. The ability to seamlessly integrate multiple agents for tasks like fundamental, technical, and market analysis showcases the potential of collaborative AI systems. How might this approach transform other industries beyond finance? https://lnkd.in/evcKaYx8 #MultiAgentSystems #AWSStrands #GenerativeAI

6

The voice AI industry has fundamentally misunderstood its own product-market fit. We're obsessing over conversational sophistication while failing at basic speech synthesis. These "performance barriers" Rishav calls out aren't technical limitations; they're the predictable result of prioritizing flashy demos over robust engineering. Companies won't adopt voice agents not because the technology isn't ready, but because vendors have consistently oversold capabilities while underdelivering on reliability. It’s about time someone said it out loud: the barriers to building RELIABLE voice AI (with stats). Here it out 👇

14

1 Comment

Building on top of large language models is fun, but getting consistent performance and reliability is extremely challenging. I love this post from Kimberly Patron about how Sierra uses health and performance-based traffic routing and request hedging to get meaningfully better performance and tail latency from high scale LLMs https://lnkd.in/g4qWRCb5

284

13 Comments