Skip to main content
Ryan Orban

Ryan Orban

Subject
7 entries

Production

Bookmarks

  1. Hive: production runtime harness for AI agents

    Hive (by Aden, YC-backed) is a production runtime harness for AI agents: state management, checkpoint-based crash recovery, cost enforcement, and self-healing through graph evolution. 102 MCP tools.

  2. Patterns for Building LLM-based Systems & Products

    Eugene Yan's comprehensive guide to patterns for building LLM-based systems — covering evals, RAG, fine-tuning, caching, guardrails, defensive UX, and user feedback collection. One of the most referenced practical engineering posts of 2023.

  3. All the Hard Stuff Nobody Talks About When Building with LLMs

    Honeycomb's post-mortem on building their LLM-powered Query Assistant — the engineering challenges they didn't expect, including output validation, latency at the tail, prompt brittleness, and user trust. Unusually honest practitioner account from a team that shipped an LLM feature to production.

  4. Running Large Language Models in Production: Cohere's TIF

    Cohere's post on their Transformer Inference Framework (TIF) — covering the systems challenges of serving large language models in production at scale. An early look at how LLM serving differs from traditional ML model serving and what optimizations matter.

  5. MLOps: Machine Learning Operations

    A comprehensive guide to MLOps — the practices, tools, and culture for deploying and maintaining machine learning models in production. Covers the full lifecycle from experiment tracking through model serving, monitoring, and retraining pipelines.

  6. Production Code for Data Science: Our Experience with Kedro

    Beamery's engineering team shares their experience using Kedro to bring software engineering discipline to data science code in production — covering what worked, what required adaptation, and how the pipeline structure changed their team's workflows.

  7. Machine Learning Engineering for the Real World

    A practical guide to ML engineering as a discipline — applying software engineering processes (agile, simplicity, iterative development) to ML projects from scoping through production. Makes the case that ML projects fail not from algorithmic complexity but from lack of engineering discipline around planning, experimentation, and deployment.

All bookmarks