Blog
Latest articles
7 min readWeaviate: Multimodal Vector Search With Built-In Vectorizers
Weaviate vectorizes text and images automatically with pluggable vectorizer modules, and combines BM25 keyword search with vector search in a single hybrid query.
5 min readLost in the Middle: Why LLMs Struggle With Long Contexts
Liu et al. 2023 showed that LLM performance on multi-document QA follows a U-shaped curve - models best recall information at the start and end of context, with severe degradation in the middle.
8 min readClaude 3 Opus: When to Pay the Premium for Anthropic's Most Capable Model
Claude 3 Opus costs $15/1M input tokens - 5x more than Sonnet. This guide breaks down exactly which tasks justify the price premium and which ones you should route to the cheaper sibling.
8 min readLiteLLM: Call 100+ LLM APIs With One Unified OpenAI Interface
LiteLLM normalizes the APIs of OpenAI, Anthropic, Bedrock, Gemini, and 100+ other providers into one OpenAI-compatible interface with built-in fallbacks and cost tracking.
7 min readCloudflare D1 + R2 + Workers: Build a Full-Stack App at the Edge for Free
D1 gives you SQLite at the edge, R2 gives you S3-compatible object storage with free egress - here is how to wire them together in a Workers app with Drizzle ORM.
5 min readNLLB-200: Meta's No Language Left Behind Translation Model
NLLB-200 provides machine translation for 200 languages including 55 low-resource African languages, with a distilled 600M model that outperforms Google Translate on 40+ languages. Practical guide with code examples and cost comparison.
7 min readOLMo 2: Allen AI's Fully Open LLM (Weights + Data + Code)
OLMo 2 is the only major LLM where you can reproduce the entire training run: weights, 3T-token Dolma dataset, training code, and evaluation suite are all public.
9 min readTechnical SEO Checklist 2026: 25 Issues That Silently Kill Your Rankings
A comprehensive 25-point technical SEO checklist covering crawlability, indexability, page speed, and structured data - with how to fix every issue.
7 min readClaude 3 Haiku: The Fastest Anthropic Model for High-Volume Production
At $0.25/1M input tokens with 200k context, Claude 3 Haiku is Anthropic's cost-optimized model. The Message Batches API cuts costs another 50%.
6 min readChroma: The Easiest Vector Database for LLM Apps
Chroma runs in-process with zero setup, embeds text automatically with a default model, and scales to a client-server deployment when you outgrow local mode.
7 min readScikit-learn in 2026: Still Relevant and What's New in v1.4+
Scikit-learn remains the best library for classical ML on tabular data - v1.4+ adds HDBSCAN, TunedThresholdClassifierCV, and better Pipeline verbosity while staying beginner-friendly.
9 min readTree of Thoughts: Let LLMs Explore Multiple Reasoning Paths
Tree of Thoughts treats problem-solving as a search problem, having the LLM generate multiple reasoning branches, self-evaluate each, and backtrack when stuck - dramatically improving on hard reasoning tasks.