Browse by Topic

A short bugfix session became a practical example of agent supervision. The agent produced valid code quickly, but it focused on symptoms first. Progress came from brief human questions that redirected work toward root cause, accurate impact framing, and a cleaner final production change set for users.

Intuitive prompt engineering - often called vibe coding - promises a flow state for software engineers, but the reality is often a repetitive loop of review and correction. We’ve traded writing code for supervising agents that write code. The question is how to make that trade worthwhile.

Intuitive prompt engineering - often called vibe coding - promises a flow state for software engineers, but the reality is often a repetitive loop of review and correction. We’ve traded writing code for supervising agents that write code. The question is how to make that trade worthwhile.

Intuitive prompt engineering - often called vibe coding - promises a flow state for software engineers, but the reality is often a repetitive loop of review and correction. We’ve traded writing code for supervising agents that write code. The question is how to make that trade worthwhile.

harrow is a thin HTTP framework built directly on Hyper 1.0. The goal is explicit, macro-free routing with built-in observability: tracing spans, Prometheus metrics, and structured request IDs from the start, not bolted on after the fact.

ro11y is a Rust observability library that implements OTLP protobuf export over HTTP without pulling in the full OpenTelemetry SDK. Seven direct dependencies instead of a hundred and twenty. It builds on the tracing crate for structured logging and distributed tracing.

marie-ssg is a static site generator built to do one thing well: turn markdown files paired with TOML metadata into HTML pages. It powers this site. The design prioritizes build speed, minimal configuration, and staying out of the way.

qrst is a hybrid search engine that runs entirely on-device. No API calls, no cloud, just a Rust binary and an embedding model. It combines BM25 keyword search (SQLite FTS5) with vector semantic search (ONNX embeddings + HNSW) and fuses results using configurable strategies including Reciprocal Rank Fusion and learned-to-rank models.

harrow is a thin HTTP framework built directly on Hyper 1.0. The goal is explicit, macro-free routing with built-in observability: tracing spans, Prometheus metrics, and structured request IDs from the start, not bolted on after the fact.

ro11y is a Rust observability library that implements OTLP protobuf export over HTTP without pulling in the full OpenTelemetry SDK. Seven direct dependencies instead of a hundred and twenty. It builds on the tracing crate for structured logging and distributed tracing.

qrst is a hybrid search engine that runs entirely on-device. No API calls, no cloud, just a Rust binary and an embedding model. It combines BM25 keyword search (SQLite FTS5) with vector semantic search (ONNX embeddings + HNSW) and fuses results using configurable strategies including Reciprocal Rank Fusion and learned-to-rank models.

harrow is a thin HTTP framework built directly on Hyper 1.0. The goal is explicit, macro-free routing with built-in observability: tracing spans, Prometheus metrics, and structured request IDs from the start, not bolted on after the fact.

marie-ssg is a static site generator built to do one thing well: turn markdown files paired with TOML metadata into HTML pages. It powers this site. The design prioritizes build speed, minimal configuration, and staying out of the way.

aws s3 sync was running at 60 KB/s on a link that should reach around 80 Mbps. The issue was not one bug but several: DFS channel pauses, Apple roaming behavior, LTE bufferbloat, and smaller configuration problems. This article walks through the complete fixes on RouterOS v7.

ro11y is a Rust observability library that implements OTLP protobuf export over HTTP without pulling in the full OpenTelemetry SDK. Seven direct dependencies instead of a hundred and twenty. It builds on the tracing crate for structured logging and distributed tracing.

A short bugfix session became a practical example of agent supervision. The agent produced valid code quickly, but it focused on symptoms first. Progress came from brief human questions that redirected work toward root cause, accurate impact framing, and a cleaner final production change set for users.

A short bugfix session became a practical example of agent supervision. The agent produced valid code quickly, but it focused on symptoms first. Progress came from brief human questions that redirected work toward root cause, accurate impact framing, and a cleaner final production change set for users.

A short bugfix session became a practical example of agent supervision. The agent produced valid code quickly, but it focused on symptoms first. Progress came from brief human questions that redirected work toward root cause, accurate impact framing, and a cleaner final production change set for users.

aws s3 sync was running at 60 KB/s on a link that should reach around 80 Mbps. The issue was not one bug but several: DFS channel pauses, Apple roaming behavior, LTE bufferbloat, and smaller configuration problems. This article walks through the complete fixes on RouterOS v7.

aws s3 sync was running at 60 KB/s on a link that should reach around 80 Mbps. The issue was not one bug but several: DFS channel pauses, Apple roaming behavior, LTE bufferbloat, and smaller configuration problems. This article walks through the complete fixes on RouterOS v7.

aws s3 sync was running at 60 KB/s on a link that should reach around 80 Mbps. The issue was not one bug but several: DFS channel pauses, Apple roaming behavior, LTE bufferbloat, and smaller configuration problems. This article walks through the complete fixes on RouterOS v7.

I built an on-device hybrid search engine that combines BM25 and vector retrieval with Reciprocal Rank Fusion. Reranking metrics suggested a learned linear fusion model would outperform RRF, but end-to-end evaluation showed otherwise. This article explains why the model matched baseline behavior and what to improve next.

qrst is a hybrid search engine that runs entirely on-device. No API calls, no cloud, just a Rust binary and an embedding model. It combines BM25 keyword search (SQLite FTS5) with vector semantic search (ONNX embeddings + HNSW) and fuses results using configurable strategies including Reciprocal Rank Fusion and learned-to-rank models.

I built an on-device hybrid search engine that combines BM25 and vector retrieval with Reciprocal Rank Fusion. Reranking metrics suggested a learned linear fusion model would outperform RRF, but end-to-end evaluation showed otherwise. This article explains why the model matched baseline behavior and what to improve next.

I built an on-device hybrid search engine that combines BM25 and vector retrieval with Reciprocal Rank Fusion. Reranking metrics suggested a learned linear fusion model would outperform RRF, but end-to-end evaluation showed otherwise. This article explains why the model matched baseline behavior and what to improve next.

I built an on-device hybrid search engine that combines BM25 and vector retrieval with Reciprocal Rank Fusion. Reranking metrics suggested a learned linear fusion model would outperform RRF, but end-to-end evaluation showed otherwise. This article explains why the model matched baseline behavior and what to improve next.

I built an on-device hybrid search engine that combines BM25 and vector retrieval with Reciprocal Rank Fusion. Reranking metrics suggested a learned linear fusion model would outperform RRF, but end-to-end evaluation showed otherwise. This article explains why the model matched baseline behavior and what to improve next.