Seattle Skeptics on AI

Tamara Weed, Sep, 12 2026

Struggling to pick the right LLM? Learn how to choose between GPT, Claude, Gemini, and Llama based on cost, scale, and specific use cases. Practical guidance for building scalable AI programs in 2026.

Tamara Weed, Sep, 11 2026

Navigate generative AI costs with our guide on build vs buy strategies, token pricing trends, and GPU infrastructure ROI. Learn how to optimize your AI budget in 2026.

Tamara Weed, Sep, 10 2026

Discover how diffusion models like Stable Diffusion amplify gender and race biases. Learn why AI image generators skew demographics and what businesses must do to mitigate risks.

Tamara Weed, Sep, 9 2026

Discover the best long-context benchmarks for LLMs in 2025. Compare LongBench Pro, InfiniteBench, and HELM to evaluate model performance on 8k to 1M+ token inputs.

Tamara Weed, Sep, 8 2026

Discover how Reinforcement Learning from Prompts (RLfP) automates prompt engineering to boost LLM accuracy by up to 10%. Learn about PRewrite, PRL, costs, and when to use this advanced technique.

Tamara Weed, Sep, 7 2026

Discover how query decomposition transforms LLM performance on complex questions. Learn why stepwise reasoning beats simple retrieval, the trade-offs in latency, and implementation strategies for enterprise search.

Tamara Weed, Sep, 6 2026

Discover how to secure Generative AI systems against unique threats like prompt injection and data poisoning. Learn essential defense-in-depth strategies, access controls, and monitoring techniques to protect your LLM applications.

Tamara Weed, Sep, 5 2026

Discover how combining pruning and quantization boosts LLM speed. Learn about HWPQ, 2:4 sparsity, and practical tips for maximizing model efficiency without losing accuracy.

Tamara Weed, Sep, 4 2026

Discover why NLP evaluation shifted from BLEU to LLM-as-a-Judge. Learn how semantic metrics and AI judges provide accurate quality assessment for modern language models.

Tamara Weed, Sep, 3 2026

Discover how vibe coding transforms natural language prompts into code architectures. Learn the risks, benefits, and best practices for using AI scaffolds without creating technical debt.

Tamara Weed, Sep, 2 2026

Discover the top transformer variants for LLMs in 2025. Learn how FlashAttention-3, Mamba, and MoE architectures boost speed and context length.

Tamara Weed, Sep, 1 2026

Discover why opinionated AI stacks outperform flexible ones in speed and retention. Learn how constrained frameworks reduce decision fatigue and boost productivity.