A complete guide to evaluating Ollama prompts with Langfuse: self-hosting Langfuse with Docker, wrapping ollama.chat with trace and generation spans that record prompts, responses, and token usage, versioning and A/B testing prompts to compare output quality across versions, recording quality scores from human raters or an automated judge model, and using the Langfuse dashboard alongside Prometheus for comprehensive AI observability.
A practical guide to generating git commit messages with Ollama: reading staged diffs with git diff –cached, a conventional commits format prompt with low temperature, a prepare-commit-msg git hook that prepends AI suggestions as editable comments, and an interactive CLI tool with accept, edit, regenerate, and quit options for full control over every commit.
A complete guide to composing Ollama with other services in Docker Compose: a full stack with Ollama, Open WebUI, FastAPI, and pgvector Postgres, a model init container, CPU-only variant, FastAPI Dockerfile and client using the compose service hostname, health checks with service_healthy dependency condition, and essential compose commands for managing models and containers.
A practical guide to email drafting with a local LLM: a basic draft_email function with tone and length parameters, subject line generation, drafting replies to existing emails with intent specification, batch personalisation of email templates for multiple recipients with company and role context, and a reference of effective tone and style prompt parameters for reliably varied email outputs.
A practical guide to LLM pretraining data pipelines: language identification with FastText, heuristic quality filtering using character-to-word ratios, symbol ratios, and repeated line detection, perplexity-based filtering with KenLM to catch templated and garbled text, MinHash LSH deduplication with datasketch, exact substring deduplication with suffix arrays, building a full pipeline with HuggingFace datatrove including Gopher and C4 quality filters, training a fastText classifier for quality scoring, and balancing the data mix across web, books, and code sources.
A complete guide to streaming Ollama token output to browser clients via WebSocket: why WebSockets suit interactive AI chat better than SSE, a FastAPI WebSocket endpoint using run_in_executor for sync Ollama, a fully async version using httpx streaming, a vanilla JS browser client with real-time token display and stop button, and a multi-client broadcast connection manager for shared AI sessions.
A practical guide to model merging for ML engineers: how linear weight averaging and model soups work, computing and applying task vectors, TIES merging with trimming and sign election to resolve conflicts between task vectors, DARE with random dropout and rescaling before merging, combining DARE with TIES for large task vectors, using mergekit with a YAML config for production merges, SLERP for smoother two-model interpolation, and a decision guide for choosing between merging methods based on task overlap and fine-tuning intensity.
A complete guide to integrating Ollama in PHP and Laravel: using openai-php/client pointed at the Ollama v1 endpoint, a Laravel AiService class for summarisation and classification, streaming SSE responses with response()->stream(), generating embeddings, and a Laravel Queue job with timeout and retry configuration for background AI processing.
A practical guide to LLM routing for ML engineers: building an embedding-based classifier router that adds under 15ms latency, generating training labels via model-agreement scoring with an LLM judge, implementing a cascade router that tries the cheap model first and escalates on low confidence, calibrating the routing threshold empirically from a quality-cost tradeoff curve, tracking cost savings versus counterfactual all-expensive routing, and deciding between a trained classifier versus cascade based on query distribution stability and labelling budget.
A complete guide to Ollama integration in Elixir and Phoenix: a clean Ollama client module using Req with pattern-matched responses, single and multi-turn chat, embedding generation, a Phoenix controller for AI endpoints, real-time streaming tokens with Phoenix LiveView using Req streaming callbacks, and background AI jobs with Oban including automatic retries.
