3 APR 2026
Library of the Week
Library of the Week — Outlines
A weekly teardown of one open-source AI/ML library: what it does, why it stands out, and when to use it.
Read this → APR 3, 2026 Office Hours
Office Hours — We're using an LLM to extract structured data from messy PDFs. Sometimes it works perfectly, sometimes it misses fields or invents data. How do I know if the problem is the model, my prompt, or the PDF quality itself?
A daily developer question about AI/LLMs, answered with a direct, opinionated take.
APR 3, 2026 The Daily Signal
The Daily Signal — April 3, 2026
Top 15 AI reads from the last 24 hours, curated from indie blogs, Substacks, and research.
Recent 12 posts
APR 2, 2026 Builders Spotlight
Builders Spotlight — DSPy
The story and philosophy behind one open-source AI project: what drove it, what makes it different, and why it matters.
APR 2, 2026 Office Hours
Office Hours — We're getting inconsistent outputs from the same prompt with GPT-5.4. Temperature is locked at 0. What's actually going on?
A daily developer question about AI/LLMs, answered with a direct, opinionated take.
APR 2, 2026 Paper of the Week
Paper of the Week — TokenPacker: Efficient Visual Projector with Group-Conditioned Dot-Product Attention for Multimodal Large Language Models...
KV cache compression that cuts memory 40–60% with under 1% accuracy loss — here's the technique your inference stack probably isn't using yet.
APR 2, 2026 The Daily Signal
The Daily Signal — April 2, 2026
Top 15 AI reads from the last 24 hours, curated from indie blogs, Substacks, and research.
APR 1, 2026 Office Hours
Office Hours — I'm using Claude Opus 4.6 for a customer-facing summarization task. Should I batch requests during off-peak hours to save money, or just call the API in real-time?
A daily developer question about AI/LLMs, answered with a direct, opinionated take.
APR 1, 2026 The Prompt Lab
The Prompt Lab — Constraint Injection
Learn the constraint injection prompting technique with concrete before/after examples.
APR 1, 2026 The Daily Signal
The Daily Signal — April 1, 2026
Top 15 AI reads from the last 24 hours, curated from indie blogs, Substacks, and research.
MAR 31, 2026 Office Hours
Office Hours — How do I know when to stop prompt engineering and just upgrade my model?
A daily developer question about AI/LLMs, answered with a direct, opinionated take.
MAR 31, 2026 The Benchmark
The Benchmark — MMLU (Massive Multitask Language Understanding)
A plain-English explainer of one AI evaluation benchmark: what it measures, how it works, and when to trust it.
MAR 31, 2026 Deep Dives
Embeddings in Practice: Every Major Model Compared
MAR 31, 2026 The Daily Signal
The Daily Signal — March 31, 2026
Top 15 AI reads from the last 24 hours, curated from indie blogs, Substacks, and research.
MAR 30, 2026 Office Hours
Office Hours — Is it better to improve the harness around the LLM or wait for a better model?
A daily developer question about AI/LLMs, answered with a direct, opinionated take.
Series
Archive 49 posts
Apr 3, 2026 Office Hours — We're using an LLM to extract structured data from messy PDFs. Sometimes it works perfectly, sometimes it misses fields or invents data. How do I know if the problem is the model, my prompt, or the PDF quality itself? Office Hours Apr 3, 2026 The Daily Signal — April 3, 2026 The Daily Signal Apr 2, 2026 Builders Spotlight — DSPy Builders Spotlight Apr 2, 2026 Office Hours — We're getting inconsistent outputs from the same prompt with GPT-5.4. Temperature is locked at 0. What's actually going on? Office Hours Apr 2, 2026 Paper of the Week — TokenPacker: Efficient Visual Projector with Group-Conditioned Dot-Product Attention for Multimodal Large Language Models... Paper of the Week Apr 2, 2026 The Daily Signal — April 2, 2026 The Daily Signal Apr 1, 2026 Office Hours — I'm using Claude Opus 4.6 for a customer-facing summarization task. Should I batch requests during off-peak hours to save money, or just call the API in real-time? Office Hours Apr 1, 2026 The Prompt Lab — Constraint Injection The Prompt Lab Apr 1, 2026 The Daily Signal — April 1, 2026 The Daily Signal
Mar 31, 2026 Office Hours — How do I know when to stop prompt engineering and just upgrade my model? Office Hours Mar 31, 2026 The Benchmark — MMLU (Massive Multitask Language Understanding) The Benchmark Mar 31, 2026 Embeddings in Practice: Every Major Model Compared Deep Dives Mar 31, 2026 The Daily Signal — March 31, 2026 The Daily Signal Mar 30, 2026 Office Hours — Is it better to improve the harness around the LLM or wait for a better model? Office Hours Mar 30, 2026 The Stack — Cursor The Stack Mar 30, 2026 The Daily Signal — March 30, 2026 The Daily Signal Mar 29, 2026 Office Hours — Should I A/B test my LLM prompts in production or is that overkill? Office Hours Mar 29, 2026 This Week in SF AI — March 29, 2026 This Week in SF AI Mar 29, 2026 The Daily Signal — March 29, 2026 The Daily Signal Mar 28, 2026 omlx: Run Local LLMs on Apple Silicon with a RAG Customer Support App Deep Dives Mar 28, 2026 The LLM Encyclopedia, March 28, 2026 LLM Encyclopedia Mar 28, 2026 Office Hours — What's the hardest part of building AI agents that actually work? Office Hours Mar 28, 2026 The Stack: Apple Silicon Local LLM Servers for Running Agents The Stack Mar 28, 2026 The Daily Signal — March 28, 2026 The Daily Signal Mar 27, 2026 Library of the Week — LangChain Library of the Week Mar 27, 2026 Office Hours — How do you actually test LLM apps beyond vibe checks? Office Hours Mar 27, 2026 Prompt Injection Prevention in Production Deep Dives Mar 27, 2026 The Inference Stack Top to Bottom Deep Dives Mar 27, 2026 The Daily Signal — March 27, 2026 The Daily Signal Mar 26, 2026 Office Hours — Why is AI agent reliability barely improving despite 18 months of model upgrades? Office Hours Mar 26, 2026 Paper of the Week — Training Language Models to Self-Correct via Reinforcement Learning Paper of the Week Mar 26, 2026 The Daily Signal — March 26, 2026 The Daily Signal Mar 25, 2026 MCP, Tool Use, and Function Calling: How Agents Actually Work in 2026 Deep Dives Mar 25, 2026 The Prompt Lab — Chain-of-Thought Prompting The Prompt Lab Mar 25, 2026 Office Hours — How are people safely reusing cached LLM answers in production RAG systems? Office Hours Mar 25, 2026 The Daily Signal — March 25, 2026 The Daily Signal Mar 24, 2026 The Daily Signal — March 24, 2026 The Daily Signal Mar 24, 2026 Office Hours — Do structured outputs from LLMs create false confidence that the response is actually correct? Office Hours Mar 23, 2026 The Daily Signal — March 23, 2026 The Daily Signal Mar 23, 2026 Office Hours — How are you handling LLM API costs in production without sacrificing quality? Office Hours Mar 22, 2026 API Rate Limits Compared: Every Major LLM Provider in One Place Deep Dives Mar 22, 2026 The Daily Signal — March 22, 2026 The Daily Signal Mar 22, 2026 Office Hours — How do I actually know if my LLM is hallucinating in production? Office Hours Mar 22, 2026 This Week in SF AI — March 22, 2026 This Week in SF AI Mar 21, 2026 The Daily Signal — March 21, 2026 The Daily Signal Mar 21, 2026 The LLM Encyclopedia, March 21, 2026 LLM Encyclopedia Mar 20, 2026 The Daily Signal — March 20, 2026 The Daily Signal Mar 17, 2026 The LLM Encyclopedia, March 17 2026 LLM Encyclopedia