Claude Code Goes No-Code

Edition 151 | January 15, 2026

The modern startup founder hiring in 2026: 8 founding engineers, zero equity dilution — all named Claude. Daily standup is asking Claude what Claude told Claude to do.

After 151 editions of Building AI Agents 🎉, we’re taking a one-week breather.

Next issue drops January 26th.

Welcome back to Building AI Agents, your biweekly guide to everything new in the field of agentic AI!

In today’s issue…

• Claude launches Cowork, giving no-code builders the power of Claude Code without the code

• Anthropic debuts HIPAA-ready healthcare agents

• New data shows 40% of agent gains are lost fixing bad outputs

• Why companies like Vercel are prioritizing feedback-driven agents

…and more

🔥 INCASE YOU MISSED IT

Readers’ favorite items from the past week

1. Amazon’s AI agent backlash, lists 180+ retailers’ products without consent

2. Google brings checkout to agents, Gemini Enterprise CX unifies shopping

3. OpenAI tests sponsored agent responses, paid placements inside chat

4. Gartner: 40% of enterprise apps will use AI agents by 2026, up from <5%

5. Microsoft’s “Agent Factory” vision, how AI teams are being rebuilt

📌 THE THURSDAY BRIEFING

Claude launches Cowork, bringing agentic AI to non-developers

Anthropic released Cowork on Monday, extending Claude Code's agentic capabilities to non-technical users through the Claude Desktop app. Users designate a folder and delegate tasks through chat: reorganizing downloads, generating expense spreadsheets from receipt screenshots, or drafting reports from scattered notes while Claude works autonomously. The tool emerged after Anthropic noticed some developers using Claude Code for everything except coding: vacation research, slide decks, email cleanup, even controlling ovens. Rather than restrict this behavior, they built Cowork in ten days, largely using Claude Code itself.

Why it matters: This puts Anthropic in direct competition with Microsoft Copilot and Google Workspace AI, and threatens dozens of startups focused on file organization and document generation. Available now as a research preview for Claude Max subscribers ($100-200/month) on macOS.

Anthropic unveils Claude for Healthcare with HIPAA-ready infrastructure

Anthropic launched Claude for Healthcare at the J.P. Morgan Healthcare Conference, days after OpenAI unveiled ChatGPT Health. For enterprises, it includes HIPAA-ready infrastructure with native integrations to the CMS Coverage Database, ICD-10 codes, the National Provider Identifier Registry, and PubMed's 35 million articles. Claude can now pull coverage requirements, cross-reference clinical guidelines against patient records, and draft prior authorization determinations for human review. For consumers, U.S. Pro and Max subscribers can connect health records through HealthEx, Function Health, Apple Health, and Android Health Connect to get plain-language explanations of lab results and medical history.

Why it matters: Anthropic's rapid response to ChatGPT Health shows they're not ceding any ground in the race for vertical dominance. But the other story is what this means for the rest of us: AI agents are quietly embedding themselves into the most personal parts of our lives. Your medical records, lab results, and insurance claims are now surfaces for autonomous AI to read, synthesize, and act on.

🤝 WITH ATTIO

Introducing the first AI-native CRM

Connect your email, and you’ll instantly get a CRM with enriched customer insights and a platform that grows with your business.

With AI at the core, Attio lets you:

• Prospect and route leads with research agents

• Get real-time insights during customer calls

• Build powerful automations for your complex workflows

Join industry leaders like Granola, Taskrabbit, Flatfile and more.

👉 Try Attio Pro for free

 🤖 AGENT OF THE WEEK

👋 Welcome back to Agent of the Week!

We've been experimenting with no-code agent builders outside of n8n, and this week I tried Lindy.ai to build something I've wanted for a while: a hiring agent that does candidate research for me.

If you've ever hired for a role, you know the pain. You post a job, get a pile of applications, and then spend hours clicking through LinkedIn profiles, Googling names, and trying to figure out who's actually worth talking to. It's tedious, time-consuming, and honestly? You start cutting corners after candidate #15.

So I built an agent that does the research and evaluation for me.

I upload a CSV of candidates (names, emails, LinkedIn URLs) and link a job description in Google Docs. The agent takes it from there. It pulls each candidate's LinkedIn profile, runs web searches to find additional context (articles, GitHub, portfolio sites), and builds a comprehensive profile for each person. Then it evaluates everyone against the job description, scores them on a 1-10 scale, and gives me a ranked shortlist with reasoning.

Spinning up simple agents like this in Lindy is fairly easy once you understand the basics of agent building.

🧠 How It Works

1. Message Received → Upload Candidates: The agent starts when you upload a CSV with candidate info (name, email, LinkedIn URL) and link your job description from Google Docs

2. Enter Loop → Process Each Candidate: The agent loops through every candidate in the list, one by one

3. LinkedIn Available? → Profile Enrichment: For each candidate, the agent checks if a LinkedIn URL was provided. If yes, it pulls their full profile data

   1. I noticed Lindy has a lot of data scrapers natively built into the platform

4. Search with Perplexity → Web Research: The agent searches the web for additional info: articles, GitHub repos, portfolio sites, anything that builds a fuller picture beyond the resume

5. Update Google Sheet → Store Results: All the enriched candidate data gets logged to a Google Sheet in real-time so you can watch it populate

6. Exit Loop → Candidate Evaluation Agent: Once all candidates are researched, an AI evaluation agent reviews the full dataset against your job description. It scores each candidate 1-10 with reasoning, flags any concerns, and produces a ranked shortlist of top candidates

7. Send Message → Deliver Results: The agent sends you a summary with the top recommendations

This agent turns manual candidate research into a simple review task. I need to customize the agent more to fit my exact needs, but this is a great starting point.

Want to learn how to build agents like this? I have full step-by-step guides inside our Building AI Agents Community.

Till next week,
✌️ AP