Week 1 Changelog: 83 Commits, 24 Features, 1 Founder

This is the week 1 changelog for SalesSheet. From February 9 to February 15, 2026, I shipped 83 commits, 24 new features, and 59 bug fixes. I am one person. No co-founder, no engineering team, no contractors. Just me, a laptop, and a fleet of AI agents.

I am building in public because I think transparency builds trust. If you are evaluating SalesSheet as your CRM, you deserve to see exactly how fast we move and what we prioritize. If you are a fellow founder, maybe this inspires you to ship faster. Either way, here is everything that happened in week one.

The Big Features

Not all 24 features are created equal. Some took an afternoon. Some took two days of focused work. Here are the ones that matter most to users:

1. AI Chat — The Core Interface

The single biggest feature of week one was getting the AI chat interface production-ready. This is not a sidebar chatbot bolted onto forms. The chat is the CRM. You type what you want in plain English, and the AI executes it — creating contacts, drafting emails, moving deals, pulling reports. Every traditional CRM action got replaced with a conversational equivalent.

2. Contact Enrichment

When you add a contact by name and email, the AI automatically enriches the record with LinkedIn data, company information, location, industry, and estimated company size. This alone saves 5–10 minutes per contact that you would spend manually researching. Over a week with 20 new contacts, that is two hours reclaimed. Read more about how this works in our contact enrichment deep dive.

3. Phone Dialer

A fully functional phone dialer built into the CRM. Click-to-call from any contact record, automatic call logging, and duration tracking. No third-party integration needed. No extra monthly fee. The dialer was one of the features that Claude Code built almost entirely on its own — I described what I wanted and reviewed the output 47 seconds later.

4. Real-time Pipeline

Drag-and-drop deal stages with real-time sync powered by Supabase. When you move a deal from "Demo" to "Proposal," every team member sees it instantly. No refresh needed. The pipeline view also supports custom stages, weighted values, and expected close dates.

5. Email Sync and AI Drafts

Two-way email sync connects your inbox to the CRM. Every email to or from a contact gets logged automatically. More importantly, the AI can draft personalized follow-ups using your deal context, contact history, and conversation tone. I covered the technical rebuild of this system in this post about our 90-test email architecture.

The Full Feature List

Here are all 24 features shipped in week one, grouped by area:

Contacts & Enrichment (6 features)

• AI contact creation via natural language chat
• LinkedIn enrichment with automatic profile matching
• Company data enrichment (size, industry, location, revenue)
• Contact search with fuzzy matching and filters
• Relationship pills showing connection strength at a glance
• Contact detail view with activity timeline

Pipeline & Deals (4 features)

• Drag-and-drop pipeline with custom stages
• Deal creation via AI chat
• Weighted pipeline values by probability
• Real-time sync across all connected users

Communication (5 features)

• Two-way email sync with Gmail and Outlook
• AI email drafting with contact context
• Phone dialer with click-to-call
• Call logging with automatic duration tracking

AI & Intelligence (5 features)

• Natural language CRM commands (the core chat interface)
• Smart model routing (fast model for lookups, capable model for drafting)
• AI-generated meeting summaries from call notes
• Deal risk scoring based on activity patterns
• Smart suggestions for next actions on stale deals

Platform & UX (4 features)

• Mobile-responsive layout with PWA support
• Dashboard with charts for pipeline health and activity
• Dark mode for the entire application
• Keyboard shortcuts for power users

59 Bug Fixes: The Unsexy Work

Features get the attention, but bug fixes are what make software actually usable. Here is a sample of what got fixed:

• Email loading flicker eliminated (covered in this post)
• Pipeline drag precision on mobile improved
• Contact search returning duplicates — fixed with deduplication logic
• AI chat responses timing out on complex queries — added streaming
• Phone dialer not logging calls under 10 seconds — threshold removed
• Dashboard charts rendering blank on first load — fixed data fetching race condition

Most of the 59 fixes were small — CSS alignment, null checks, edge cases in data parsing. But each one made the product a little more polished. Death by a thousand cuts works in reverse too: a thousand small fixes create a product that just feels right.

How AI Agents Made This Possible

Let me be direct: shipping 24 features in one week as a solo founder is not normal. Five years ago it would have been impossible. Today it is my operating model, and the reason is AI coding agents.

My workflow looks like this: I describe a feature in natural language. Claude Code reads my entire codebase, understands the architecture, generates the implementation, and writes tests. I review the output, make adjustments, and merge. The average feature goes from idea to merged code in under an hour. Some take 10 minutes.

This does not mean the AI does everything. I still make every product decision, design every interaction, review every line of code, and fix the subtle bugs that require understanding user intent. The AI handles the mechanical part — the typing, the boilerplate, the CSS, the test scaffolding. I handle the creative part — the what and the why.

"The AI is the hands. I am the brain. Together we move at 10x the speed of either alone."

What Is Coming in Week 2

Week two is focused on three themes: team collaboration features, Slack integration, and analytics depth. If you want to follow along, I will be publishing another changelog next Sunday. You can also check out The AI-Native Solo Founder Playbook for a deeper look at how this operating model works.

If you are a sales team looking for a CRM that ships features at startup speed with the polish of a mature product, give SalesSheet a try. The product you see today will be meaningfully better next week. And the week after that. That is the advantage of a solo founder with AI agents — zero bureaucracy, maximum velocity.