In 2018, I started my first job as a developer. Fresh out of the gate, eager to learn, and ready to take on anything. One of the first things handed to me? A chatbot system.
At Talkpush, a recruitment SaaS company, businesses used the platform to streamline hiring with a CRM and AI-powered chatbots. These weren’t just one-off bots—it was a chatbot system that allowed companies to deploy their own recruiter or interviewer chatbots on platforms like Messenger and WhatsApp. The bots could screen candidates, ask pre-qualifying questions, and collect responses, automating a huge part of the recruitment process.
The chatbot system was already built when I joined. My job wasn’t to create it from scratch but to take ownership—maintain it, optimize it, and improve its functionality. That was my first real experience with AI-driven automation, and I was immediately hooked.
The Struggles and the Magic of Early AI
Back then, AI wasn’t what it is today. There were no Large Language Models (LLMs) like we have now. Instead, the chatbot system relied on state machines and NLP engines like Dialogflow.
Each bot followed a structured conversation flow. Users had to give expected responses for the conversation to continue. If a candidate answered in an unexpected way, things could break fast. Training the NLP models meant manually adding every possible variation of a question to make sure the chatbot understood different ways people might respond.
It was both frustrating and fascinating. On one hand, it was amazing to see how AI could automate conversations at scale, reducing human effort in recruitment. On the other hand, there were clear limitations. If a bot didn’t understand, it couldn’t just "figure it out" like AI does today.
Even with all the challenges, I loved working on it. It wasn’t just about writing code. It was about designing an intuitive experience, making AI actually useful in a business setting, and seeing how automation could transform an industry.
Fast Forward to Today – And Why I’m Building Fororo
Now, five years later, I’m diving back into AI-driven automation. But this time, I’m not working on chatbots for recruitment. I’m tackling something that affects all knowledge workers: productivity.
Over the years, I’ve noticed a growing problem. We’re all drowning in tasks, juggling priorities, and constantly switching between tools. We don’t have a productivity problem. We have a focus problem.
That’s why I’m building Fororo—an AI-powered productivity assistant designed to help knowledge workers stay focused, manage tasks effectively, and cut through digital distractions.
Unlike traditional task managers that just act as to-do lists, Fororo actually helps users decide what to work on next. It does this by analyzing urgency, deadlines, and work patterns to suggest tasks based on impact and efficiency.
What Makes Fororo Different?
✅ AI-Powered Task Prioritization – No more decision fatigue. Fororo ranks tasks based on urgency, importance, and personal work habits.
⏳ Web-Based Self-Timer – Tracks time spent on tasks, helping users stay accountable and build better focus habits.
📊 Productivity Insights Dashboard – Visualizes work patterns and distractions, giving users actionable insights to improve efficiency.
Full Circle
It feels like I’ve come full circle. The same excitement I had in 2018—exploring AI and automation—is back. But this time, instead of automating recruitment workflows, I’m using AI to help professionals work smarter, not harder.
Fororo is my way of taking everything I’ve learned—from state machines to NLP to LLMs—and building a tool that actually makes an impact.
I’ll be sharing more about Fororo’s development soon. If you’ve ever felt overwhelmed by tasks or struggled to focus, stay tuned.
Ja じゃ 👋🏼