- Operators
- Posts
- 014 AI For Operators
014 AI For Operators
Jaron Sander, Growth Engineer at Inform Growth; Review of Bardeen.ai, 10 links
Hi there,
Welcome back to AI for Operators. Here’s what we’ve got for you this week:
The Operator: Jaron Sander, Founder & Growth Engineer at Inform Growth (listen to the episode here)
The Review: Bardeen.ai - a workflow automation tool with native web scraping capabilities.
The Links: 10 links, including multiple perspectives on AI Agents, lessons from enterprise AI adoption, and an essay debating whether evals are important.
The Operator | ![]() |
Jaron Sander, Founder & Growth Engineer at Inform Growth ![]() | ![]() |
In this episode, Jaron shares how he built an Asana-native events agent to automate repetitive planning while keeping humans in the loop, including. Some of his insights include:
Investing time in scoping and planning is always worth it - it helps you understand the problem, figure out what actually needs to be built, and shows you where automation can help.
Prototype, then productize: Jaron built a working prototype in in Make.com to validate the workflows, then migrated costly steps to a scheduled Python script once usage and value were clear.
Build proprietary databases and other context-rich inputs wherever possible: the events agent relies on databases of validated speakers, previously-used venues, and more, which means that the outputs are more predictable and less costly.
Keeping humans in the loop ensures quality: agents aren’t really capable of true end-to-end execution yet, especially on critical, public-facing tasks. Having lightweight checkpoints along the way ensures hallucinations are caught and the result is high quality.
Generalizable playbook: build inside existing tools, add explicit rulebooks and checkpoints, and aim for a 50–70% assist so people focus on higher-impact decisions.
The Review | ![]() |

This is not a sponsored post.
What it Does
Bardeen is a workflow automation tool with native web scraping capabilities, making it particularly effective for operators who want to run automations in their browser and as opportunities come up (versus setting up complex automations once and having them trigger more regularly).
Why Ops Leaders Should Care
Finding the perfect workflow automation tool is tough. And Bardeen isn’t perfect - at least not for every use case. But if you or your team’s work involves lead generation and follow-up, monitoring websites, updating spreadsheets, repetitive recruiting work, or transferring data between systems, Bardeen could be a good way to save a lot of time.
Key Features (Pros & Cons)
Pros
User-friendly, no-code interface makes automation simple and accessible for even the least technical operators.
Automations and workflows, especially around data entry tasks, can help operators save significant amounts of time.
Web scraping and browser automation helps Bardeen stand out from other no code automation tools, which can be particularly helpful for customer-facing teams that need to pull live information from customers’ or prospects’ sites.
Privacy-conscious: all automations run locally in your browser, which means sensitive data isn’t uploaded to Bardeen’s servers.
Strong, positive user feedback on G2 and Product Hunt.
Cons
Bardeen’s approach to ‘proactive automation’ means that users need to trigger workflows rather than them happening automatically in the background like Zapier. This has advantages, but also has disadvantages.
Like most workflow and automation platforms, there’s a fairly steep learning curve that requires dedicated time and attention for you to take full advantage of the product.
Usage-based pricing means that it can get pricey quickly if you’re a power user.
While Bardeen has hundreds of integrations, if you’re using a particularly niche tool, the team may not have built that integration yet, which may make it less useful for you.
Other Options

Bottom Line
Bardeen is a powerful tool great for non-technical users (and teams). While the usage-based pricing means it can get pricey fast, folks with lightweight use cases (or those just wanting to try it out) can still save a lot of time without spending a bunch of money. And after all, if you’re burning through your credits by automating a ton of tasks across different platforms and workflows, it may be a sign that you should make the transition from a no-code tool like Bardeen to hybrid or more code-first solution - or at least that’s one of the lessons we learned in this week’s podcast with Jaron.

The Links | ![]() |
Practical
GPT-5 - The Case of the Missing Agent: Reflections on the capabilities (and lack of capabilities) of tools currently labeled as ‘AI agents’. We’ve still got a long way to go, and it’s important to be aware of what ‘agents’ can do and what they can’t (yet).
Scaling Agents Internally: Lessons from Atlassian
Agentic Best Practices: A guide on how to build agentic systems.
Model Personalities: Do different models have distinct personalities? You bet.
Perspectives
Are Evals Important? The CTO and Co-Founder of Raindrop reviews what an eval is and the best arguments for why they’re valuable.
Lessons from Enterprise AI Adoption: Consumers & prosumers are AI-obsessed, but the evidence on AI adoption (and successes) is mixed. What’s working and what’s not?
Anthropic’s Agent Framework: As usual, Anthropic is focused on making sure agents are safe and trustworthy.
News
Is AI taking jobs? Noah Smith dives into the complex economic data.
How NetStock Gets SMBs to adopt AI: Bottom’s up adoption and fitting into existing workflows is key.
More Enterprise Adoption Case Studies: How Block and GlaxoSmithKline grafted AI onto existing workflows to drive adoption.

Thanks for reading,
Tom Guthrie