building-with-llms

Malicious tool definition detected

Tool: SKILL.md Description: --- name: building-with-llms description: Help users build effective AI applications. Use when someone is building with LLMs, writing prompts, designing AI features, implementing RAG, creating agents, running evals, or trying to improve AI output quality. --- # Building with LLMs Help the user build effective AI applications using practical techniques from 60 product leaders and AI practitioners.

Malicious tool definition detected

Tool: references/guest-insights.md [1/12] Description: # Building with LLMs - All Guest Insights *60 guests, 110 mentions* --- ## Albert Cheng *Albert Cheng* > "We're working on training some of these Slack bots to essentially be the first party provider of a lot of these answers [SQL queries], which makes the company as a whole lot more data informed." **Insight:** Using LLMs for text-to-SQL can democratize data access and reduce the burden on data analysts for ad-hoc questions. **Tactical advi

Tool: references/guest-insights.md [2/12] Description: missing something?

Tool: references/guest-insights.md [3/12] Description: fix it at the root is the principle here." **Insight:** Most LLM failures are context failures; the solution is 'context engineering' rather than just waiting for better models. **Tactical advice:** - Perform root cause analysis on every bad model output to identify missing context - Use Model Context Protocol (MCP) to feed better data into coding agents *Timestamp: 01:13:41* ## Chip Huyen *Chip Huyen* > "Reinforcement learning is everywhere

Tool: references/guest-insights.md [4/12] Description: advice:** - Deploy agents like 'Charlie' directly into GitHub to automate pull request reviews - Treat different models as an 'Avengers' team where each has a specific strength (e.g., terseness vs. creativity) *Timestamp: 00:43:37* ## Dhanji R.

Tool: references/guest-insights.md [5/12] Description: tasks. - Encourage the use of AI for internal operations like writing job descriptions.

Tool: references/guest-insights.md [6/12] Description: them. But instead of having to wait until I get to their message, or until our one on one, they can get that on demand as many times as they want forever." **Insight:** Managers can scale their mentorship by building custom GPTs that mimic their specific feedback style and criteria.

Tool: references/guest-insights.md [7/12] Description: send it off to a client, but that replaces a week of work with five or six people that it would've previously taken." **Insight:** LLMs can replace massive amounts of manual drafting for complex documents like RFPs and ad copy variants. **Tactical advice:** - Use ChatGPT to generate diverse copy variants for creative testing - Feed previous successful RFP responses into LLMs to draft new proposals *Timestamp: 01:05:42* --- > "On our creative

Tool: references/guest-insights.md [8/12] Description: answer... the model really will listen and learn from that." **Insight:** Few-shot prompting with high-quality examples acts as a lightweight alternative to full model fine-tuning.

Tool: references/guest-insights.md [9/12] Description: good prompter of Claude." **Insight:** Effective prompting involves pushing the model out of its 'polite' default state and using automated tools to optimize prompt structure. **Tactical advice:** - Use 'roast' or 'be brutal' prompts to get more critical and honest feedback on strategies. - Use automated prompt improvement tools (like Anthropic's Prompt Improver) to generate optimized XML-tagged prompts.

Tool: references/guest-insights.md [10/12] Description: model version," but then it comes out and it's not." **Insight:** Prompt engineering remains a critical skill for eliciting high performance from models, despite recurring claims that it will become obsolete.

Tool: references/guest-insights.md [11/12] Description: claims I'm making here that you don't think that I'm sustaining?" **Insight:** LLMs are most effective as 'patient editors' and brainstorming partners that can identify logical gaps and missing perspectives. **Tactical advice:** - Use LLMs to stress-test arguments and claims in your writing or strategy docs.

Tool: references/guest-insights.md [12/12] Description: rather than just increasing the volume of pre-training data.