Friday 29th May’s issue is presented by Fidelity Careers

Sharpen Your Skills. Elevate Your Career.

We know working in tech is about more than just shipping code - it’s about fostering a sustainable craft and building for the long term. Fidelity technologists thrive in a culture focused on curiosity and professional reinvention:

• 💡 Curiosity: encouragement to imagine and build the next generation of tech

• 📚 Growth: support to learn, advance, and master new skills

• 🚀 Versatility: tools and support to reinvent your career

Lead Tech Forward At Fidelity

Get To The Core Of The Thing

— Shreyas Doshi

tl;dr: “In every product conversation, the framing decides the discussion. People rise to whatever level of abstraction the question opens up. Board conversations especially. But it’s the same with executive staff conversations and team conversations.”

Leadership Product Communication

Be A Thermostat, Not A Thermometer

— Lara Hogan

tl;dr: We are easily influenced by the mood of those around us — “one person’s behavior change can cause others to change their behavior,” and by setting the whole temperature for the room, they’re being a thermostat. As leaders, Lara advises us to pick up on these negative mood changes early and become the person that sets the the new temperature of the room in a positive and healthy way. She illustrates how to do so here.

Management Communication

Build The Future Of Tech At Fidelity

tl;dr: At Fidelity, we’re a community of technologists committed to boldly building the next generation of fintech—and helping you advance. By investing in your technical skills, supporting training and education opportunities, and encouraging career reinvention, we empower you to follow your curiosity and grow a tech career you love. Learn more about growing your tech career at Fidelity and visit Tech.FidelityCareers.com

Promoted by Fidelity Careers

CareerGrowth

When Too Many Maps Overlap On One Person

— Yusuf Aytas

tl;dr: “You have a proposal. And someone suggested checking with Mike. It does not sound like a problem. It sounds like due diligence, like someone being careful, like a team that knows its own limits. And in the moment, it usually is all of those things. But if you have seen this before, you start to notice what happens right after it gets said. The proposal does not stop. Nobody declares it blocked. The meeting continues. But something is different, and everyone feels it, and nobody mentions it, because what would you even say?”

Leadership OrgDesign DecisionMaking

Beyond The Prompt: Claude Code

— Arpan Patel

tl;dr: “The casual user types prompts, accepts suggestions, and treats it like a fancier autocomplete. The daily driver uses it like a programmable agent with memory, custom commands, parallel sessions, and a project setup that compounds over time. This guide is for the second kind of person, assuming you already know what Claude does when you type it in a terminal.”

Guide Agents AI

Training An Agentic Search Model To Be 20x Faster Than GPT-5

— Sam Dauncey, Max Rumpf

tl;dr: Frontier LLMs are incredible generalists, but expensive specialists. As token usage comes under scrutiny, there’s an emerging trend toward small models trained for specific tasks. SID-1 is an agentic search model that runs complex retrieval tasks 20x faster than frontier LLMs with 99% fewer tokens - at double the recall of traditional RAG pipelines. Here’s how it was trained.

Promoted by turbopuffer

AI DeepDive

Cheap Turpentine

— David Singleton

tl;dr: “There’s a whole genre of writing about company building that reads like art criticism — frameworks, principles, taxonomies of culture. Useful, sometimes. But the conversations I find myself actually wanting to have with other founders are the turpentine ones: what’s your offer letter template, what does someone’s first day look like, how do you run a standup that doesn’t suck, what do you do when you run out of space around the lunch table?” David shares the actual assets he used at his company.

Management EarlyStage

How's Linear So Fast? A Technical Breakdown

— Dennis Brotzky

tl;dr: “A few milliseconds is all it takes to update an issue in Linear. A traditional CRUD app doing the same thing takes about 300ms. How do they do it? There's no secret silver bullet to performance. The reality is that it's built from the ground up on the right foundation, then improved by countless decisions. My goal is to walk through some of the techniques that make Linear feel the way it does and help you implement the same.”

Architecture Performance DeepDive

Using AI To Write Better Code More Slowly

— Nolan Lawson

tl;dr: “A lot of people seem convinced that the point of AI coding is to write low-quality code as fast as possible. Spew out barely-passable slop, open massive PRs, and merge them unvetted. Ship it! But the thing is, LLMs are very flexible. And you can use them just as effectively to write high-quality code more slowly.”

DevEx CodeReview AI

Null Pointer

Hand-drawn by Manu. View the Null Pointer series

Most Popular From Last Issue

Agentic Engineering Patterns - Simon Willison

Notable Links

Agent Governance: Policy enforcement, identity, sandboxing & SRE for AI agents.

Cate: A spatial desktop IDE.

Honker: Queues, streams & time-trigger scheduling inside your SQLite file.

OpenSpec: Lightweight framework for spec-driven development.

Tools: Small, low stakes and low effort tools.