- Founder With AI
- Posts
- The Real Work Often Feels Boring
The Real Work Often Feels Boring
Burnout, MicroSaaS ideas, and the value of showing up anyway
Section 1: Behind the Scenes of My Week
Writing, Niche Research, and the Quiet Burnout
This week was quieter on the writing front.
I posted less than usual. But surprisingly, X grew faster than last week mostly because of a lead magnet I made around Veo 3. It was trending, so I put together a quick guide. It worked. Got good engagement on X. Not much on LinkedIn, but still a small win.
No big calls or external moments this time. Just a growing internal realisation.
This journey is not going to be easy.
I started diving deeper into niche problems. Not just general pain points but specific issues faced by a small set of users. That took up most of my time and energy this week.
That’s how I ended up exploring Linear, the product management tool.
I noticed that many teams still create release notes manually. Some even write custom workflows to automate it.
So I had a thought what if an AI agent could do that for them?
I shared the idea in their community. A few users responded. Nothing flashy, but the feeling was different. For the first time in a while, I felt like I was circling a real problem.
Still, the path ahead is unclear. I need to figure out the right way to package and monetise it. But the early signal felt good.
Alongside all this, I’ve been feeling the quiet kind of burnout.
Some days I’m writing. Some days I’m deep in research. Some days I’m trying to keep projects moving or support the team.
And some days, I’m not sure what to focus on.
But I’ve come to accept that part too. This is not about momentum every day. It’s about staying in the game long enough to let the pieces click.
Even slow weeks move the story forward.
Section 2: AI Workflow of the Week → Personalized LinkedIn Replies at Scale
This week’s workflow was built for a common but underrated use case — replying to inbound LinkedIn messages with context and care, without spending hours in the inbox.
The challenge:
→ You get messages asking about your product or service
→ You want to reply thoughtfully
→ But you’re either too busy or the messages pile up and go stale
So I built a solution using n8n, OpenAI GPT-4o, Notion, and LangChain.
Here’s how it works:
→ Input Reception and Data Isolation
The workflow is triggered by a parent process (like a webhook or CRM entry). It grabs the message content and sender info from LinkedIn.
→ Context Enrichment via Notion
It queries a Notion database that holds different message types and preferred response guidelines. This provides context like:
What type of inquiry it is (pricing, partnership, general interest)
What tone or structure we prefer for that kind of reply
→ AI Response Generation with LangChain Agent
Using GPT-4o and session memory, it generates a personalized reply that feels warm and relevant — not generic. It factors in the sender’s profile and the response rules from Notion.
Sample response:
“Hey Ankit, appreciate you reaching out. If you’re exploring AI automations for recruiting, I’d be happy to share what we’ve built recently. Let me know a good time to connect.”
Why this matters:
→ Helps you reply fast without sounding robotic
→ Keeps the tone consistent across team members
→ Reduces cognitive load when handling a busy inbox
📖 View the Full Step-by-Step Guide with Workflow JSON
Section 3: Problem Spotlight → The Quiet Pain of Release Notes
While researching niche SaaS problems this week, I spent some time inside Linear — a beautifully designed project management tool used by a lot of fast-moving teams.
One pattern stood out.
Many teams were still creating release notes manually.
Some had built custom scripts. Others were just copy-pasting issue titles. A few skipped it entirely because it was too much effort.
The process looks something like this:
→ Mark issues as done
→ Open a doc or Notion page
→ Summarize updates manually
→ Clean it up and send it to the team or post in changelog
That’s hours of work every week — just to summarize what was already done.
So I started exploring a fix.
→ Track completed issues in Linear
→ Extract title, labels, and context
→ Prompt an AI model to write a clean, categorized summary
→ Format it for internal docs, changelogs, or public updates
Why this is worth solving:
→ Writing release notes is important but easy to delay
→ Developers rarely enjoy the task
→ Even a partial automation here could save time and improve consistency
I shared the idea in Linear’s community and got a few early nods. Nothing flashy — but it felt like I was close to something real.
Still exploring how to bundle this into a useful agent. Monetization might be tricky, but the pain is clear.
If you’re on a product team and hate writing release notes, I’d love to hear how you currently handle it.
This might be a quiet problem worth solving.
Section 4: Tool Highlight → Kittl. An AI Design Sidekick
This week I stumbled on Kittl, a Berlin‑based design platform that feels like a light version of Canva with AI built in .
Why it stood out
Kittl packs a powerful set of AI-enhanced design utilities into one browser-based platform. You’ll find tools for:
→ Logo generation
→ Background removal and upscaling
→ Vector creation and editing
→ Typography and layout suggestions
It offers a free tier, no signup required for basic use—which makes it easy to test and integrate into your work.
How I used it
I needed quick social post visuals this week and ran them through Kittl’s AI image generator and background remover. It took less than a minute to get a clean, branded asset that didn’t feel generic.
Why you should care
→ Keeps your design workflow in one tab
→ Helps non-designers create polished assets fast
→ Free enough to test without friction
If you spend time crafting visuals but don’t want to learn Photoshop or switch tools all the time, Kittl is worth a look.
Let me know if you try it!
Section 5: Open Loop → A Quiet Week Still Counts
This week reminded me of something I often forget.
You don’t need explosive results every week to make progress.
I didn’t post much. No leads came in. No big wins.
But a few things still moved:
→ I researched real problems
→ Got a few signals from a community post
→ Built one useful internal tool
→ Showed up even when I felt off
And that matters.
Not every week will bring momentum. But every week is a chance to keep the signal alive.
So here’s a question I’ve been asking myself:
👉 What’s one quiet problem your audience deals with daily — but never posts about?
That’s where your next product might be hiding.
You don’t have to rush. You just have to look.
If you’re on the same path building, writing, testing ideas I’d love to hear what you’re circling around right now.
Just reply. I read every message.
Reply