AI agents that can automate your work using your data, built to help humans move faster. Connect your tools, describe a task, and agents do the reporting, cleaning, writing, and updating for and with you.
No complexity. No workflows. Just agents.
Today, we’re excited to launch Calk AI 1.0, the platform we’ve been building quietly between Canada 🇨🇦 and France 🇫🇷 - with one mission:
Bring AI to teams without complex workflows, mapping, or technical barriers.
In 2025, AI for businesses is stuck between two extremes:
As teams 😩
AI assistants can’t access company data
Tools that promise simple automation but require workflow building
Too many AI tools that don’t talk to each other
Workflow tools require technical skills
AI that feels powerful… until it breaks
As founders/operators 😤
Drowning in vertical AI tools
Still doing repetitive tasks manually
Too much scattered data
ChatGPT is amazing but blind
Workflows feel like engineering
We saw companies everywhere struggle with the same thing:
AI sounds easy. But deploying it is still too damn hard.
So we built a new way.
👾 What is Calk AI ?
AI agents that work with your real data - without workflows.
🔌 Connect Your Tools — live data every time 🧠 Build Agents in 3 (fun) Steps — no workflows needed ⚡ Push Actions — update your tools automatically 📅 Scheduling — automate daily, weekly, monthly 🧩 Agent Superpowers — extend capabilities instantly ⚡ (Soon) Triggers — react to events instantly
🌍 Who we are
Built by two founders, bootstrapped from day 1, obsessed with simplicity
We’re a tiny team of two - one in France, one in Canada (cc @harryf) - building fast, shipping every week, and listening closely to the early users who believe in agents as the next wave of automation.
We want to build a tool that empowers humans, not replaces them.
Agents that help people get more done with simplicity, data, and speed.
@mohsinproduct Thanks for the question! Not a generic failure! the agent will tell who the problem and it will try it again before telling you this! For exemple, if your agent is connected to Airtable Hubspot and let's say Gmail. If the agent can't connect itself to gmail, it will tell you what the problem is. They are self aware of what they do!
@mskyow thanks for the kind words ! It’s simple as describing what you want to get done. You describe the agent then let ai handle the prompting part. Then connect the tools, and yes it can handle multi steps process ! Try it and tell us !
@axel_lorier Thanks Axel! 🙌
Easiest way to compare:
n8n = you build workflows (nodes, logic, branches).
Calk = no workflows at all. You just describe the task, connect your tools, and the AI agent handles the logic.
Example: daily CRM update
• n8n → 10+ nodes
• Calk → 1 agent, 3-step setup
So basically:
n8n is for building workflows.
Calk is for people who don’t want to build workflows.
Happy to explain more!
Report
@quentin_fournier_martin Love this direction, most tools still expect users to build flows, map fields, or design automations before anything useful happens. Skipping all that and letting agents work directly off your connected data feels like the closest thing yet to “real” workplace automation.
Curious: what types of tasks people are getting the most value from so far, reporting, data cleanup, or more creative work like writing?
@fernando_scharnick Thanks Fernando, really appreciate it 🙏
You nailed it. Most tools still force people to build flows before anything useful happens.
Calk lets agents act directly on your connected data - zero silos, zero configs.
The biggest wins we’re seeing so far:
1️⃣ Reporting & insights — weekly snapshots, campaign reviews for internal of clients purposes
2️⃣ Personal task automation — inbox, CRM, tasks, all the “I shouldn’t be doing this” work
3️⃣ Data flow between tools — HubSpot ↔️ Notion ↔️ Airtable without building a single workflow
Happy to share real agent examples if you want!
Report
@quentin_fournier_martin That’s super helpful, appreciate you breaking down the top wins. The “no workflows, just outcomes” approach really shows in those use cases, especially the cross-tool data flow. That alone usually takes hours of setup in other systems.
And yes, would love to see a few real agent examples whenever you’re able to share, always interesting to see how people are using this in the wild!
@abod_rehman thanks ! We strongly believe that AI agents are still too complicated to build & Deploy! On your side how do you build AI agents for internal purposes on Triforce ?
Replies
Calk AI
Hey Product Hunt 👋
Today, we’re excited to launch Calk AI 1.0, the platform we’ve been building quietly between Canada 🇨🇦 and France 🇫🇷 - with one mission:
Bring AI to teams without complex workflows, mapping, or technical barriers.
In 2025, AI for businesses is stuck between two extremes:
As teams 😩
AI assistants can’t access company data
Tools that promise simple automation but require workflow building
Too many AI tools that don’t talk to each other
Workflow tools require technical skills
AI that feels powerful… until it breaks
As founders/operators 😤
Drowning in vertical AI tools
Still doing repetitive tasks manually
Too much scattered data
ChatGPT is amazing but blind
Workflows feel like engineering
We saw companies everywhere struggle with the same thing:
AI sounds easy. But deploying it is still too damn hard.
So we built a new way.
👾 What is Calk AI ?
AI agents that work with your real data - without workflows.
🔌 Connect Your Tools — live data every time
🧠 Build Agents in 3 (fun) Steps — no workflows needed
⚡ Push Actions — update your tools automatically
📅 Scheduling — automate daily, weekly, monthly
🧩 Agent Superpowers — extend capabilities instantly
⚡ (Soon) Triggers — react to events instantly
🌍 Who we are
Built by two founders, bootstrapped from day 1, obsessed with simplicity
We’re a tiny team of two - one in France, one in Canada (cc @harryf) - building fast, shipping every week, and listening closely to the early users who believe in agents as the next wave of automation.
We want to build a tool that empowers humans, not replaces them.
Agents that help people get more done with simplicity, data, and speed.
🤝 Thanks to our hunter, @fmerian 🙏
Flo has been incredibly supportive - thanks for helping us launch today!
We’ll be here all day answering questions, taking feedback, and improving Calk AI with the PH community 💛
Tell us what agents you want built - we’ll probably ship them next week.
Let’s build the future of (simple & human) AI automation together 🚀
PicWish
@harryf @fmerian @quentin_fournier_martin congrats!
If an agent fails a step in a multi steps, do we get a log or just a generic failure notification?
Calk AI
@mohsinproduct Thanks for the question! Not a generic failure! the agent will tell who the problem and it will try it again before telling you this! For exemple, if your agent is connected to Airtable Hubspot and let's say Gmail. If the agent can't connect itself to gmail, it will tell you what the problem is. They are self aware of what they do!
Swytchcode
This is wonderful. Congrats on the launch
Calk AI
@chilarai Thanks! Hope you'll enjoy using Calk!
Calk AI
@chilarai Thank you!
BeeDone
nice product, it's cool
Calk AI
@layouceferie Thanks! Any bad experience on automation on n8n so far ?
Calk AI
@layouceferie Thanks! Don't hesitate to tell us your experiences with AI agents so far!
No-code AI agents that actually understand your business? This is what we've been waiting for! 🔥
How complex can the workflows get? Can these agents handle multi-step processes?
Calk AI
Calk AI
@mskyow Thanks Vlad,
Can our agents handle multi-step process? The answer is yes!
You give it access to the tools you want and it will orchestrate the functions within it to execute to correct actions.
Depending on what you want to do, if you explain it well, it will be able to understand and act on it!
Adjust Page Brightness - Smart Control
This is called gold!
Calk AI
@kshitij_mishra4 Wow! Thanks for this man!
Calk AI
@kshitij_mishra4 Thank you so much! Do you struggle to build agents too?
That is awesome! Do you have an integration with n8n or Cargo? (Congrats by the way!)
Calk AI
@sanjay_lalsingue1 Soon for n8n! And Cargo we can think about it! @maxence_de_villepion can we integrate Cargo to our solution ? 🤓
Guideflow
Calk AI
@clement_puren2 Thanks! Clément! Huge help from Guideflow to do this launch!
Congrats on the launch! How do you compare Calk to n8n ? I don't get it
Calk AI
@quentin_fournier_martin Love this direction, most tools still expect users to build flows, map fields, or design automations before anything useful happens. Skipping all that and letting agents work directly off your connected data feels like the closest thing yet to “real” workplace automation.
Curious: what types of tasks people are getting the most value from so far, reporting, data cleanup, or more creative work like writing?
Calk AI
@quentin_fournier_martin @fernando_scharnick Hi Fernando, thanks for the kind words.
The tasks can be varied but the most useful ones are about getting data from various sources from only one agent and then be able to iterate on it.
Not having to spend 20 minutes searching for data is a great feeling, once that's done, you can go a be as creative as you want with it.
Calk AI
@quentin_fournier_martin That’s super helpful, appreciate you breaking down the top wins. The “no workflows, just outcomes” approach really shows in those use cases, especially the cross-tool data flow. That alone usually takes hours of setup in other systems.
And yes, would love to see a few real agent examples whenever you’re able to share, always interesting to see how people are using this in the wild!
Triforce Todos
You’ve solved a real pain point. Big congratulations on shipping Calk AI 1.0!
Calk AI
@abod_rehman thanks ! We strongly believe that AI agents are still too complicated to build & Deploy! On your side how do you build AI agents for internal purposes on Triforce ?