Articles The hardest part of building software is not coding, it’s requirements Lessons from Bootstrapped Companies Founded by Software Engineers Risk Entity Watch — Using Anomaly Detection to Fight Fraud ML Pipeline Architecture Design Patterns Tutorials How it works: The novel HTTP/2 ‘Rapid Reset’ DDoS attack Tools three-glob, WebGL Globe Data Visualization as…

Articles 4 lessons from launching an LLM chatbot Finding gaps in the knowledge base is the surest way to improve a RAG app Evaluate the RAG app end to end using GPT-4 A/B test to evaluate different prompt strategies Evaluate retrieval in isolation Generative AI’s Act Two 7 Habits of Highly…

Articles How DoorDash Fosters Meaningful Engineering Career Development Should I Rust or should I Go? Reversible and Irreversible Decisions When decisions are reversible, make them fast When decisions are irreversible, slow them down Can LLMs Really Reason and Plan? nothing that I have read, verified or done gives me any compelling…

Articles Measuring developer productivity? A response to McKinsey Scaling the Instagram Explore recommendations system What I Have Changed My Mind About in Software Development self-documenting code unit testing private methods using an IDE using a debugger working remotely using chatGPT Tutorials Understanding and Mitigating LLM Hallucinations Tools numaflow, Kubernetes-native platform to run massively…

Articles How to build an enterprise LLM application: Lessons from GitHub Copilot

Articles Costs exposed: On-call ticket handling a clear “on” / “off” split for operational support; a clear process to transfer ongoing incidents to the following on-call person, along with an expectation that this is the normal thing to do; and a clear distinction between mitigation practices and resolution efforts. whenever a…

Articles Cell-based Architecture on AWS

Articles How to Build a Fully Automated Data Drift Detection Pipeline

Articles Open challenges in LLM research 1. Reduce and measure hallucinations 2. Optimize context length and context construction 3. Incorporate other data modalities 4. Make LLMs faster and cheaper 5. Design a new model architecture 6. Develop GPU alternatives 7. Make agents usable 8. Improve learning from human preference 9. Improve the efficiency of the chat interface 10. Build LLMs…