Field Signal · Curated Reading
The Engineering
Signal.
High-signal field notes, architecture teardowns, and engineering writing from 50 publications. A focused reading queue for becoming a stronger end-to-end engineer. Updated 19 Aug.
The New American AI Model Designed to be Customized
In this article, we will work through the various choices Thinking Machines made while building Inkling.
Fragments: August 18
Part of the reason why I’m at Thoughtworks is because I’d like to see a software development organization founded on technical excellence as an example for the rest of the industry. The trouble is that I have little aptitude or inclination for the hard work of building such an organization. So I rel...
Consistency is the new latency: AI at the data layer
As AI agents move from chatbots to taking action, their reliability depends on the consistency of the data layer beneath them. This post examines how replication lag poisons an agent's context and shows how to match Amazon Aurora, Amazon DynamoDB, and Amazon Keyspaces replication models to each task...
Waymo vs Tesla: Two Ways to Build Self-Driving Cars
In this article, we will take a look at both approaches.
EP222: What is Google’s TPU?
A TPU (Tensor Processing Unit) is Google’s custom AI chip, designed from scratch for the giant matrix multiplications that modern models live on. GPUs were built for graphics first.
Serverless vehicle tracking at scale: Bosch L.OS on AWS
Learn how Bosch Mobility Platform Solutions built L.OS, a serverless vehicle tracking platform on AWS that unifies India's fragmented spot logistics market into a single real-time visibility layer using Amazon ECS, AWS Lambda, and Amazon MSK.
Track generative AI costs with Amazon Bedrock inference profiles
Learn how to track generative AI costs by department using Amazon Bedrock application inference profiles and AWS cost allocation tags. Create tagged profiles for each team and view per-department cost breakdowns in AWS Cost Explorer.
A Detailed Guide to API Composition Techniques
In this article, we are going to dive deep into the area of the API composition problem and the patterns associated with it.
Recovery strategies to meet data residency requirements
Learn three strategies for achieving disaster recovery while meeting data residency requirements. Ranging from encryption-based controls on multi-Region replication to fully in-country architectures, these patterns help you balance recovery objectives with regulatory constraints.
Reducing Text2SQL latency with parameterized query templates
Learn how parameterized query templates reduced Text2SQL latency by 80% and cut token consumption by over 50%. This post covers the architecture behind an intelligent caching layer that uses semantic similarity to match user questions to SQL templates, bypassing expensive LLM calls.
Adobe Firefly: Simplified observability with Amazon Managed Prometheus
Learn how Adobe Firefly achieved 28x faster GPU metric queries by migrating from self-managed Prometheus to Amazon Managed Service for Prometheus, with improvements in query performance, infrastructure reliability, and operational efficiency.
GitHub vs Vercel vs Replit: What Dev Platforms Do When AI Code Is Cheap
AI models have solved the writing code part of software development to a great extent.
Scaling patterns for self-organizing multi-agent clusters with Kiro
Learn how to coordinate AI agents through shared state in Amazon S3 instead of a central orchestrator. Deploy and observe self-organizing agent clusters on Amazon EC2 with the open-source kiro-flock reference implementation.
How Cloudflare Is Making AI Pay for Content
In this article, we will go through Cloudflare’s solution in the following five steps.
TDD inside the agent loop - theater or actual value?
My colleagues at Thoughtworks tend to be big fans of Test-Driven Development, and many people in the industry advocate telling LLM agents to use TDD when building software. Birgitta Böckeler was curious if this really makes a difference, so conducted a few experiments. more…
How to Fight Clickbait: Meta, LinkedIn & YouTube Case Studies
What does it take for a social media platform to stop rewarding clickbait content?
The Read Path versus the Write Path: Strategies and Techniques
In this article, we will look at read path and write path operations and techniques in detail.
How Big Models Teach Small Models to Be Smart
In this article, we will walk through the idea from the ground up.
The Pulse: Bending Spoons' Acquisition Strategy
In only 5 years, Hopin went from zero to a $7.7B valuation, and back to zero again. Also: Bending Spoons’ startup acquisition model.
Why An LLM’s Memory Gets Expensive and How to Fix It
In this article, we will learn how LLMs use memory, how it gets expensive, and how to fix it.
Fragments: August 4
There’s been a fair bit of publicity of the Open AI “rogue agent” that hacked into Hugging Face. This prompted Anthropic to check what their models were up to and, to my complete lack of surprise, discovered three incidents where models had gained unauthorized access to data in other organizations. ...
LLM Security Basics: The Full Threat Model
In this article, we try to build a map of the full attack surface that threatens an LLM’s security.
Hiring: Part Time Instructor, Write Production Grade Code with AI
We’re hiring a part-time instructor for “𝐖𝐫𝐢𝐭𝐞 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐆𝐫𝐚𝐝𝐞 𝐂𝐨𝐝𝐞 𝐰𝐢𝐭...
The Conductor Developer
TL;DR Why I think software development is starting to feel a little more like conducting an orchestra. There’s a shift happening in software development that I don’t think we’re talking about clearly enough. For the last couple of years we’ve framed AI as a productivity tool. How much faster can it ...
The Pulse: Quitting Spotify Podcasts over reliability
Spotify’s podcast platform has become chronically unreliable since the company’s leadership started boasting about AI adoption. Competitors haven’t had similar issues, so I offboarded from Spotify.