How Claude Code Transforms Team Workflows

Anthropic published a detailed look at how its own teams use Claude Code. It might not be applicable in full for every company but the results are still worth reading carefully. The expected findings were there. Engineers moving faster. Debugging time down. More experiments running in parallel. The unexpected findings were more interesting. Lawyers built internal phone tree systems to route team members to the right counsel. Marketers generated hundreds of ad variations in seconds. Data scientists created visualizations without knowing JavaScript. Finance staff described what they needed in plain language and received working Excel outputs. These were people with no programming background solving real operational problems by describing them. The underlying dynamic is structural. Most organizations ration software development through a backlog. Teams with ideas wait for engineering capacity. The constraint shapes what gets built and what gets dropped. When anyone who can describe a problem can build a solution, the backlog changes shape. The new bottleneck becomes problem definition. How clearly a team can articulate what they need. Anthropic's own teams are a live test of what happens when that shift occurs across an organization. The published research documents it in detail. For enterprise buyers: the organizations moving fastest on this are asking a different question. Whether their people can describe what they need clearly enough to build it. Link to the report is in the comments.

Claude Code can now build HubSpot workflows. Not describe them. Build them. HubSpot quietly shipped a Workflows API in beta. That means AI agents can now create workflows, not just trigger them. This is another big addition to the Claude <> HubSpot stack. And this one is big for admins, RevOps, and solution partners. Before: Claude could read your data, update records, and start workflows. But building the automation was still your Tuesday afternoon. Now: Claude Code can read your portal, spot the gap, and create the workflow through the API. You QA it in the UI. You flip the toggle. The flow: - Agent analyzes patterns and process gaps in the portal - Proposes an automation - Builds the workflow via the API - Admin QAs and turns it on - Agent tests and refines based on performance Nothing here replaces the human. The admin still decides what "good" looks like. The solution partner still owns the architecture. But the "sit down and build 14 workflows this sprint" part? That just got compressed. For HubSpot Solution Partners, this also shifts delivery economics. Clicking through the workflow builder is no longer the main cost. The value moves to strategy, QA, and ongoing governance. HubSpot keeps calling its vision the "agentic customer platform." Every release like this is another piece of that scaffolding. Claude Code can talk to HubSpot via MCP and API. Now it can build inside it too. Start working with Claude <> HubSpot today. You won't just be early. You'll be ready when HubSpot gives you more power in the future. P.S. Would you let Claude build a workflow in your production portal today, or sandbox only? #HubSpot #Claude #RevOps #API

🎯 The Developer Is Now The Orchestra Conductor Four weeks ago, as I became familiar with Claude Code and adopted it as the coding assistant of choice, I came to realize that its evolution would fundamentally shift my role from hands-on-keyboard pair-programmer to agent manager. Possibly, orchestra conductor. This week, July 25 proved that prediction right—Anthropic's official sub-agents launch just made multi-agent development workflows production-ready … almost overnight. 🔧 What I'm seeing in practice: The DEVELOPER → REVIEWER → VERIFIER → GIT-MANAGER process of development workspace compliance I've been refining is now officially supported. Instead of co-authoring code, I'm designing agent personalities. ⚡ The technical breakthrough: Separate context windows per agent have solved the coordination nightmare. • No more context pollution • No more community workarounds • Just clean, specialized AI teams working in parallel 💡 Here's what most miss: This isn't about replacing developers—it's about elevating the developer who can think like an architect and manage the development process. I spend my time now on: ▶ Architecture decisions ▶ Quality gates ▶ Strategic orchestration Meanwhile, my agent fleet handles implementation details. The cognitive load has shifted from syntax to systems thinking. 📊 Real numbers: Anthropic's own teams process hundreds of code additions in minutes using specialized sub-agents. Their dev teams run autonomous loops—code, test, iterate—with human oversight at commit points. 🎯 The nuanced reality: Human involvement is still critical. Someone needs to design the agent personalities, manage the handoffs, and maintain quality standards. That someone is the developer who understands both code and coordination. We're not coding less; we're architecting more. The future belongs to developers who master agent orchestration, not those clinging to individual contribution. Lest anyone consider this a slight on the incredible, cutting-edge work of Reuven Cohen, let me counter that sustained success delivering production code using frameworks like claude-flow, requires the kind of depth of knowledge, experience and skills he and others like Adrian Cockcroft bring to the party. 🔮 What's next?: Within months, job descriptions will shift from "senior developer" to "senior agent-based development manager." The question isn't whether you can code — it's whether you can think in terms of design patterns and architecture, then incorporate your skills in agent management for high-speed software development. Are you ready to put down the keyboard and pick up the conductor's baton? 🎼 #ArtificialIntelligence #TechLeadership #SoftwareDevelopment #SoftwareDevelopment #MultiAgentSystems

*** How the Creator of Claude Code Actually Uses It *** Boris Cherny just shared his Claude Code workflow, and it’s surprisingly practical. Key takeaways for technical leaders: 1/ Setup Philosophy He says it works great out of box with minimal customization. The tool is intentionally flexible and every team member uses it differently. 2/ Scale & Parallelization He runs 5 local Claude instances + 5-10 web sessions simultaneously. Distributes work across terminal tabs and devices (desktop + mobile), using system notifications to track progress. 3/ Model Selection Uses Opus 4.5 exclusively for coding. Despite being slower, it is a superior tool and less steering is required which makes it faster end-to-end than smaller models. 4/ Team Knowledge Management Shared CLAUDE.md file in git tracks coding standards and common mistakes. When Claude makes errors, they’re added to prevent repetition. Compounding Engineering in practice. 5/ Workflow Automation ∙ Starts with Plan mode (shift+tab twice) before executing ∙ Custom slash commands for repeated workflows (commit-push-pr, typecheck, lint) ∙ Subagents handle post-work tasks (code-simplifier, verify-app) ∙ PostToolUse hooks auto-format code to prevent CI failures 6/ Security & Permissions Pre allows safe bash commands via /permissions to eliminate friction. No skip permissions flag. 7/ Tool Integration Claude accesses Slack (MCP), BigQuery, Sentry logs, and other internal tools directly. Configuration checked into .mcp.json and shared across team. 8/ Quality Multiplier Most critical insight I got was to give Claude ways to verify its own work. Adding feedback loops (testing, validation) improves output quality 2-3x. To me his posts are somewhat counter intuitive as he talks of using more plain vanilla settings, uses the slower Opus 4.5 and uses fleet of agents not just a coding assistant. Which I guess is why I learned so much from it! https://lnkd.in/egveGqzh

The moment you find a way to give agentic AI like Claude Code access to all of your corporate tools and context, everything changes. New AI drops every week. Cerebras, Opus 4.6, OpenClaw, Moltbook. I'm sure several more launched while I was writing this post! I'm overwhelmed trying to keep up while actually trying to stay on top of my job. So here's one concrete thing that's actually changed how I work. The problem with every ChatGPT-style tool is that you are the integration layer. You copy from Jira, paste into the chatbox. Copy from Slack, paste into the chatbox. The AI is smart, but it's totally blind. You're doing all the research so the AI can do the thinking. So... a couple weeks ago I started using Claude Code with MCP connections to my actual work tools — Gmail, Calendar, Slack, Jira, Google Docs, Snowflake, etc — and something clicked. It's not just that it has context (though that matters a lot). It's that it's actually agentic. It doesn't just answer questions — it takes actions. It searches, cross-references, drafts, creates (with your permission, of course!). It operates across your systems instead of waiting for you to spoon-feed it. I asked Claude: "What's the current state of the current migration project? Check Jira, Slack, and the design doc." And of course, Claude just did the work: searched my real systems. Pulled threads together on its own, no 15-minute research manual context gathering phase before I could even start thinking. The bottleneck hasn't been model intelligence for a while now! It's context and agency. Once the agent can see your systems and act on them, the whole workflow inverts. You stop doing the work and start directing and refining the work. "Draft a follow-up email to yesterday's design review — pull the notes from the doc and the attendee list from the calendar invite." It does the research, writes the draft, you ask for revisions, make your edits, remove the still-too-real AI slop effect as much as you can, and send. That shift matters most for managers. Our days can sometimes feel like 70% or more context-gathering and 30% judgment calls. AI that can't see our systems only helps with the 30%. An actual agent eliminates the entire gathering phase. There's a million AI tools fighting for your attention right now. Ignore most of them. Connect one agent to your real work systems and see what happens.

Claude Code did not just grow fast. It became a warning signal. From zero to $1B in revenue in six months. Possibly already closer to $2B. Faster than ChatGPT reached that milestone. But the real story is not the revenue. It is who is using it. Microsoft engineers are reportedly using Claude Code internally, even though Microsoft sells GitHub Copilot. A Google principal engineer said Claude reproduced a year of architectural work in one hour. At Epic, the healthcare technology company behind MyChart, more than half of Claude Code usage reportedly comes from non-developer roles. Support teams. Implementation teams. People who were never supposed to “code.” Novo Nordisk used it to cut regulatory documentation from more than 10 weeks to 10 minutes. And the person prototyping new features was not a software engineer. They had a PhD in molecular biology. That is the shift. AI is no longer only making developers faster. It is turning domain experts into builders. 70% of Fortune 100 companies are now using Claude. Engineering teams are reporting 2x to 10x speed improvements. This is not a trend anymore. It is becoming the new baseline. For me, this is the urgent lesson for every leader: The future of work will not be divided between “technical” and “non-technical” people. It will be divided between people who can turn expertise into systems, and people who cannot. The question is no longer: “Should my team use AI?” The question is: “Which parts of my organization are already being outpaced by teams that do?” What role do you think changes the most when non-technical experts can suddenly build with AI? #ArtificialIntelligence #ClaudeCode #Anthropic #AI #FutureOfWork #SoftwareEngineering #DigitalTransformation #Leadership #Automation #EnterpriseAI

Last week I asked PMs whether they use Claude Code. A third of them didn't know where to start. And when I read the comments, even PMs who do use it are overwhelmed by how fast everything moves and unsure which workflows are worth investing in. Every week there's a new list of "must-use" agents and skill packs, and the pressure to overhaul your setup is real. But most PMs are still trying to figure out what fits their actual work and what's just noise. Pulling signals from bugs, feature requests, competitor moves, VP asks, user research. Stitching it together. Formatting it. Making it presentable. The actual product thinking gets whatever time is left. These are the kind of problems that skills in Claude Code are ideal to solve. A skill in Claude Code is a workflow you describe once and run with one command. You don't need a hundred. You need three or four that solve your specific week. Three of my favorite that anyone can use to learn about /skills: /research Pull a weekly AI research brief every Monday. Three developments, one quote, one action item. Used to take 45 minutes of tab-switching. Now it runs while you make coffee. /meeting-prep (name) If you connect searches Gmail for shared docs, open action items, and recent threads. Returns the three things to cover in your next meeting. Run it 10 minutes before every 1:1. /prd-review This one is fun. Run a PRD through three parallel agents, each reading from a different stakeholder persona (eng manager, director, PM peer). Three independent gap lists before the doc goes to anyone. Pro tip: ask your boss and your eng manager to help you build a persona that resembles them, then your PRD reviews can be much more effective. Building skills in Claude Code is very easy. Each one started as a conversation where I described the problem, refined the output until it matched what I wanted, and saved it. And the best part is that they can build on each other. You can have, for example, the research skill feed into meeting prep. One workflow becomes the input to the next. The data gathering, sorting, wrangling and eternal scrolling work still happens. It just doesn't have to eat your calendar anymore. I wrote a step-by-step guide on building /skills from scratch this week on Product SideQuest and I believe everyone, especially Product Managers, should be building skills and delegate the 'busy' work to agents. #ProductManagement #AI

Most think Claude Cowork = Claude with folder access. That barely scratches the surface. The real power of Cowork isn’t the task you type. It’s what happens before and after you type it. Cowork actually runs as a 4-layer execution system: 1️⃣ Global Instructions → Your permanent identity Claude knows who you are before you say a word. Tone, role, format, preferences — loaded every session. 2️⃣ Folder Instructions → Project-level context Drop into any folder and Claude already knows the rules. Brand voice, output format, naming conventions — all pre-loaded. 3️⃣ Plugins → Domain expertise on demand Not just a generalist anymore. Finance, HR, legal, ops — each plugin changes what Claude knows. 4️⃣ Sub-agents → Parallel execution Complex tasks get broken into workstreams. Multiple Claude instances running concurrently. You just wait for the output. Most people open Cowork and start describing tasks. But the gap between a rough output and something near-final almost always comes down to setup. Practices that actually move the needle: → Write global instructions before session one — treat it like an onboarding doc → Add folder instructions per project — don’t re-brief Claude every time → Install role-specific plugins before you start a task → Use persistent agent threads for work that spans multiple days → Schedule recurring tasks — weekly reports, daily briefs, file processing — set once, runs forever Go deeper: 67 use cases organised by profession (actually wild) https://lnkd.in/gaq7vWRF 10 copy-paste workflows that work today https://lnkd.in/g5szdbNh Complete 2026 setup + prompting guide https://lnkd.in/grK3NfSu Official getting started docs https://lnkd.in/g_4kXFFD Bookmark this if you’re a knowledge worker building with Claude. The future of work isn’t using AI harder. It’s building a system that works without you.

Claude Code's new Agent Teams feature is the most significant leap in AI-assisted coding since Claude Code itself. And I've been testing it heavily over the past few days. Most people hear "multiple agents" and think subagents. Subagents are isolated workers. They go off, do a task, and report results back to the main agent. No coordination between them. If your frontend subagent makes a decision that affects the backend, the backend subagent has no idea. Agent teams are fundamentally different. They share a task list with dependency tracking and message each other directly. Peer-to-peer, not hub-and-spoke. When my frontend agent decided on an API contract, it messaged the backend agent directly. The backend agent adapted in real time. I had 3 agents build a full Claude Agent SDK orchestrator app. Frontend agent set up the React UI, backend agent wired up the API, database agent designed the schema. They negotiated interface contracts with each other without me directing traffic. 6 minutes of wall clock time for what would've taken a single agent 20+. Anthropic's own proof point: 16 agents built a 100,000-line Rust C compiler that compiles the Linux kernel. $20,000 in API costs, ~2,000 sessions, 99% pass rate on GCC torture tests. Not a toy demo. It's not perfect yet. Token cost runs 2-4x a single session since each agent maintains its own context window. The lead agent sometimes implements things itself instead of delegating (you have to steer it). No session resumption if something crashes. It's still a research preview with rough edges. But the coordination working at all is the breakthrough. This is the difference between hiring 3 freelancers who never talk to each other versus 3 teammates in a room. No other coding tool has this right now. I built a skill to handle task decomposition and team structure upfront, which makes the delegation way more predictable. I just posted a full breakdown on YouTube, link here: https://lnkd.in/gzcHW-He And here is a link to the Agent Team skill in GitHub: https://lnkd.in/gWPRA6Mj

Most people use Claude Code like a chatbot. They type a prompt. Wait for a response. Copy paste the output. Repeat. That's not how Claude Code works. Claude Code is an agent. It lives on your machine. It reads your files, runs your terminal, remembers your project, and connects to your tools. Here's what that actually means: 𝗙𝗶𝗹𝗲𝘀 → Your file system is Claude's workspace. You don't upload anything. It already sees everything. It can organize your folders, read 50 docs, summarize them, and move files around. 𝗖𝗼𝗺𝗺𝗮𝗻𝗱𝘀 → Your terminal is Claude's hands. It runs code, fails, reads the error, and fixes itself. You describe the task in plain English. It executes. 𝗠𝗲𝗺𝗼𝗿𝘆 → Claude remembers your project across sessions. Your preferences, your past decisions, your context. Every session builds on the last one. 𝗧𝗼𝗼𝗹𝘀 → MCP servers let Claude talk to Gmail, Slack, Notion, Jira, databases. You can extend it with custom skills for any workflow you run repeatedly. You don't prompt Claude Code. You orchestrate it. Vibe coding builds apps from prompts. Claude Code runs your entire workflow. Vibe coding lives in the browser. Claude Code lives on your machine. Vibe coding starts fresh every project. Claude Code remembers across sessions. Most people confuse the two. They're not even close. Over to you: Are you using Claude Code as a chatbot or as an agent?