Autonomous AI Agents Move From Concept To Business Backbone
AI automation agents are software systems that can independently understand business context, decide on the next step, and carry out tasks across tools without humans manually triggering each action. This shift from static dashboards to autonomous business systems is drawing fresh investor attention, as shown by two recent Series A rounds worth a combined USD 55 million (approx. RM253 million). Airspeed is building an “execution layer” for revenue teams, while Lassie focuses on automating administrative work for small businesses. Both target the same trend: software that does the work instead of only reporting on it. Yet their market segments, go-to-market strategies, and products are very different, hinting at how large and varied the emerging category of agent-native platforms could become.
Airspeed: An Agent-Native Execution Layer For Revenue Teams
Airspeed has raised €17.2 million (USD 20 million, approx. RM92 million) in Series A funding to expand its agent-native platform for go-to-market execution. Founded in 2022 by former DeepMind research scientists Adam Liska and Devang Agrawal, the company pitches itself as the “system of action” that revenue teams are missing between systems of record and systems of intelligence. Its AI automation agents work across calls, emails, tickets, and CRM data to update records, flag risks, and generate follow-ups that move deals forward. According to Airspeed, it operates as a commercial brain that unifies context, memory, and knowledge across the revenue workflow, while keeping humans in control. The company reports 200 customers in 20 countries and says monthly agent run volumes nearly tripled from January to April, as customers created thousands of custom agents.
Lassie: Autonomous Admin For Small Businesses, Starting With Healthcare
Lassie has secured USD 35 million (approx. RM161 million) in Series A financing to build autonomous business systems for small companies. Founded by former product leaders from Robinhood, Coinbase, and Superhuman, Lassie targets the heavy administrative burden that weighs on owners and staff. Its small business AI tools are already in use at more than 700 businesses across 49 states, providing over 250,000 hours of labor each year. The initial focus is healthcare practices, where an AI agent logs into insurance portals, pulls reimbursement data, reconciles records, updates systems of record, and checks deposits in bank accounts. Lassie says a typical medical practice can lose over 100 hours per month to such tasks and spend about USD 200,000 (approx. RM920,000) annually on related staffing, making revenue team automation-style gains compelling far beyond the enterprise.
Different Markets, Same Bet: Agents That Do The Work
Despite different customers, Airspeed and Lassie both signal confidence that AI automation agents are the next step in business software. Airspeed targets enterprise and growth-stage revenue teams that already use complex CRM stacks and want deeper revenue team automation. Lassie, by contrast, serves small businesses that lack large ops teams and need autonomous business systems to reclaim time from back-office work. Investor interest from firms such as DN Capital and Andreessen Horowitz suggests a belief that software categories will split between traditional tools and agent-native platforms that can act on live data. Airspeed focuses on GTM execution across the customer journey, while Lassie concentrates on repeatable, rules-heavy admin workflows like insurance reimbursements. Together, they show how agent-native platforms are forming a distinct category beyond legacy automation scripts and templates.
The Rise Of Agent-Native Platforms As A New Software Category
Both companies position themselves as early examples of agent-native platforms, which differ from classic workflow automation in several ways. Instead of simple triggers and rules, these systems keep a persistent understanding of context, update that understanding in real time, and act with guardrails. Airspeed describes its stack as a unified commercial model, an agent runtime, and rigorous evaluation to keep every action trustworthy. Lassie emphasizes repeatable, outcome-focused workflows, where hours of manual work per medical practice each month can be shifted to AI automation agents. As tools evolve from assisting users to partially running operations, questions around reliability, oversight, and role design will decide who benefits most. For now, investors are betting that the winners will be platforms where agents are not add-ons, but the core of the product.
