Code w/ Claude — Talk Series
An Instagram carousel by Charlie Hills (charliehills.substack.com) cataloguing the Code w/ Claude talk series (London, UK). 14-slide carousel; talks captured below with runtimes.
Note: This page was originally bookmarked under
@noevarner.ai, but the linked post resolves to Charlie Hills’ “Code w/ Claude” series. Attribution corrected to match the actual post.
Talks
- Running an AI-native engineering org — Fiona Fung (Mgr, Claude Code & Cowork Team, Anthropic) · 26:38 — How Anthropic’s own engineers ship now that writing code is the easy part. Review is the new bottleneck.
- Getting more out of the Claude Platform · 26:41 — Prompt caching, tool search, compaction, and the advisor strategy, demonstrated live on the Claude API.
- Designing with Claude: From prompt to production · 28:00 — Plain-language brief → brand-locked, production-quality design output. No Figma needed.
- Building with Claude on Google Cloud · 24:38 — A feedback app built on Google Cloud using subagents and MCP in 30 live minutes.
- Memory and dreaming for self-learning agents · 21:34 — Memory architecture plus dream consolidation — the pattern that lets agents improve between runs with no retraining.
- What legal agents inherit from coding agents: Lessons from Legora · 28:43 — How coding-agent patterns translate to legal work: document editing, contract linting, bulk review.
- How to get to production faster with Claude Managed Agents — Michael Cohen (Anthropic) · 29:04 — Anthropic handles the infra, state, and permissioning so you ship the agent, not the plumbing.
- Build a production-ready agent with Claude Managed Agents (workshop) · 27:23 — Deploy a production agent, then debug it live inside the developer console.
- Building AI-native at enterprise scale: monday.com, Doctolib, Delivery Hero · 29:10 — Three European unicorns picked Claude. Where each broke first and how the team fixed it. (Ruslan Semenov / monday.com; Alex Kaluzny, CTO / Doctolib; Rodrigue Schäfer, VP Platform / Delivery Hero)
- From one person to 80: Scaling a hypergrowth engineering org with Claude Code — Gabriel Grinberg (AI Eng Manager, Base44) · 23:58 — Why keeping the architecture boring is what let Claude Code scale the team alongside the product.
- Build a proactive agent workflow with Claude Code · 22:03 — Run routines that read your repo and open PRs before you even open your laptop.
- Build AI agents using Claude in Microsoft Foundry (workshop) · 34:16 — Provision Claude inside Foundry, wire up MCP, and ship a working agent end-to-end.
- Coding is no longer the constraint: Scaling devex to teams and agents at Spotify — Niklas Gustavsson (Chief Architect & VP Eng) · 27:36 — 96% of Spotify devs code with AI. The Honk agent and the guardrails that accelerate output instead of slowing it.
- AI with Claude on AWS: From code to orchestration · 19:42 — Claude Code on Bedrock in 45 live minutes. CLAUDE.md plus skills tuned for your team’s dev environment.