Hey Makers
The wait is over! We are incredibly proud to announce the official finalists for our Community Challenge: Designing Intelligent Workflows with Make AI Agents!
Over the past few weeks, we watched this community redefine what automation with AI can look like when you put a reasoning engine at the center. From solo-operator assistants to multi-agent operating systems, the quality and ambition of the submissions left us speechless. The sheer number of submissions alone was one our highest ever!
Narrowing down to a shortlist was incredibly difficult. But as with any contest, whether we like it or not, we had to eventually narrow it down. So, without further ado, here are the twelve finalists heading into the final round:
DISC Lead Personaliser 🎯 by @Dan_M
The Solution: An emotional intelligence infrastructure that upgrades AI agents from basic sentiment analysis to human-level nuance. By combining Make with a custom MCP server, it classifies social media comments into 12 distinct emotions in real-time, allowing agents to detect true frustration (even when a customer says they are “fine”) and identify high-value leads.
The Win: Ends “tone-deaf” automation by providing AI with granular, queryable emotional data. It saves agencies 15+ hours a week, eliminates manual monitoring anxiety, and turns messy human emotions into quantifiable business metrics and automated escalation flows.
Check out the video for this solution here: DISC Lead Personaliser.
WhatsApp Booking Agent📱 by @Itay_Lavi
The Solution: A conversational AI agent embedded in WhatsApp that handles the full booking lifecycle for small service businesses — scheduling appointments, processing reschedules, and sending follow-ups — entirely autonomously, around the clock.
The Win: Gives micro-businesses a 24/7 front desk without the cost of one. No more missed bookings outside business hours, no more manual back-and-forth over availability.
Check out the video for this solution here: WhatsApp Booking Agent.
Make Security Auditor 🔐 by @Annas_Almukahal
The Solution: A security testing agent that ingests your Make scenarios, fires real attack payloads against them, and returns a confirmed list of vulnerabilities — not theoretical ones.
The Win: Moves security from a checkbox exercise to an active, evidence-based practice. It finds the gaps that manual reviews miss and gives builders a concrete remediation list.
Check out the video for this solution here: Make Security Auditor.
Legal Agentic OS ⚖️ by @NeillHuman
The Solution: A full-stack legal operations layer built from specialist AI agents that collaborate to handle client intake, document review, matter triage, and escalation — operating as a coordinated automated team rather than a single bot.
The Win: Compresses the administrative overhead of a legal practice into an autonomous operating layer. It routes the right work to the right place instantly, so human lawyers spend time on law, not logistics.
Check out the video for this solution here: Legal Agentic OS.
Make Error Doctor 🩺 by @Mariia_Bondar
The Solution: A diagnostic agent that activates the moment a Make scenario fails — it reads the error, identifies the root cause, selects the appropriate remediation action, and delivers a precise fix-it briefing to exactly the right person.
The Win: Turns cryptic failure notifications into actionable intelligence. No more digging through logs or guessing what broke — the agent does the forensics and hands a solution to the human in plain language.
Check out the video for this solution here: Make Error Doctor.
Solopreneur AI Assistant 🧠 by @Marco_Handke
The Solution: A coordinated family of AI agents that proactively manages a solopreneur’s full operational stack — email, calendar, task list, and CRM — without waiting to be asked. It monitors context, anticipates needs, and takes action.
The Win: Gives the one-person business the operational leverage of a full-time EA. It removes the cognitive overhead of context-switching and keeps the founder focused on the work that only they can do.
Check out the video for this solution here: Solopreneur AI Assistant.
Conversational Sales Agent 💬 by @Firas_RIDENE
The Solution: A Slack-based agent that takes a lead, researches them autonomously, scrapes their website, assembles a personalised pitch deck, and queues the outreach for human approval — all from a single trigger.
The Win: Compresses hours of pre-sales research and deck preparation into minutes. The sales rep stays in the loop for the final call but is handed everything they need, fully prepared.
Check out the video for this solution here: Conversational Sales Agent.
AI Governance Roadmap 🗺️ by @Daraima_Bassey_N
The Solution: An 8-agent system that analyses a company’s current AI tool profile and generates a practical, risk-calibrated governance implementation roadmap — scaled to the organisation’s actual exposure level, not a generic template.
The Win: Makes AI governance accessible to companies that don’t have a dedicated compliance team. It translates a complex, ambiguous challenge into a sequenced, actionable plan.
Check out the video for this solution here: AI Governance Roadmap.
Farmer Knowledge Engine🌾 by @Monica
The Solution: A private, personalised search engine that answers domain questions in the owner’s voice, drawing from their own accumulated expertise — and built with deliberate protections against content scraping.
The Win: Solves the “expert knowledge trapped in one person’s head” problem while protecting the IP that makes that expertise valuable. The farmer’s knowledge becomes searchable, scalable, and secure.
Check out the video for this solution here: Farmer Knowledge Engine.
Autonomous Ops Brain⚡by @Pramod_Sahu
The Solution: An operations agent that continuously monitors for customer issues, instantly pulls relevant data from across connected systems, and either resolves the issue autonomously or escalates it — in seconds, not hours.
The Win: Eliminates the lag between a customer problem appearing and someone acting on it. The agent closes the loop faster than any human queue can, and does it with full system context.
Self-Healing Support 🔄 by @MRodoreda
The Solution: A support agent that resolves tickets from an existing knowledge base — and when it encounters a gap, automatically generates the missing documentation so the same question never goes unanswered a second time.
The Win: Turns every unanswered question into a permanent improvement. The support system gets smarter with every ticket, compounding its own value over time without manual content curation.
Content Strategy Pivot 📊 by @Joshua_Peter
The Solution: A daily content agent that reads its own performance analytics, makes a data-driven decision on whether to stick with the content plan or pivot, writes platform-specific posts, and publishes a full reasoning log explaining its choices.
The Win: Closes the loop between content performance and content creation — automatically. No more weekly reporting meetings to decide what to post next. The agent reads the data, reasons through it, and executes.
Check out the video for this solution here: Content Strategy Pivot.
Notable Mentions
We also want to highlight two submissions that stayed with us after judging.
Our absolute team favourite was a solution by @ApexSystematic. The scenario monitors an elderly parent via WhatsApp, detects signs of distress in their messages, tracks their location, and sends the family a daily well-being summary. It’s the kind of build that reminds you what automation is actually for. And although it maybe wasn’t technically as strong as some of the others, our unanimous reaction was to give it a major shout-out and explore what we can do with it further.
And the second mention goes to @Dirk_Baard — for building what we’ve agreed is the most creative submission in the challenge. His scenario fully automates the creation of collectible trading cards for a website called MB Says. Whilst this solution doesn’t necessarily solve a pain point, but it absolutely demonstrates the builder’s skills, and we respect the commitment.
What happens next? Our finalists move into the final round, where our Community Champions will select the Grand Winner — announced on June 16th!
To everyone who participated: Thank you. Building something real takes courage, and every submission here showed it. We hope to see you in the next challenge.
Who are you rooting for? Drop your pick in the comments! 