The countdown is on! Service Summit 2025 is just around the corner, and we have an exciting session you won't want to miss. Join Rasa's Chief Product Officer, Björn Minkmar, for "Conversations You Can Count On". 📅 Date: November 20, 2025 🕚 Time: 11:30 AM - 12:00 PM 📍 Location: Mont Blanc Stage In this session, Bjorn will reveal how Rasa transforms the promise of AI agents into high-trust, resilient systems that are truly worthy of your customers.
About us
The Rasa Platform is the leading open generative conversational AI platform to build and deliver next-level AI assistants. With Rasa, brands can leverage the power of generative AI to create adaptive brand experiences for their customers. By automating processes, improving customer satisfaction, and gathering valuable insights, Rasa enables brands to create exceptional customer interactions while reducing costs. Trusted by startups to Fortune 500 companies, Rasa ensures data privacy, security, and scalability for enterprises of all sizes. Rasa is privately held with funding from Accel, Andreessen Horowitz, Basis Set Ventures, and others. The company was founded in 2016 and is remote-first with a presence in San Francisco, Berlin, London, Paris and Belgrade.
- Website
-
https://rasa.com
External link for Rasa
- Industry
- Software Development
- Company size
- 51-200 employees
- Headquarters
- San Francisco, California
- Type
- Privately Held
- Founded
- 2016
- Specialties
- natural language processing, dialogue management, NLP, NLU, AI, conversational AI, machine learning, bots, chatbots, assistants, deep learning, Conversational AI, Conversational AI Platform, AI Assistants, and Virtual Agents
Locations
-
Primary
521 7th St
San Francisco, California 94103, US
-
Schönhauser Allee 175
Berlin, Berlin 10119, DE
Employees at Rasa
Updates
-
Your agent’s architecture is its attack surface. We partnered with Lakera, an independent AI security firm. The research, conducted independently by Lakera, tested two agents under the same conditions: 🔹 Same functionality 🔹 Different designs 🔹 Same attacks 1️⃣ What we learned: Most teams double down on prompts + guardrails. But it’s the architecture that decides what’s even possible to exploit. 2️⃣ In the prompt-driven agent, a fake “debug mode” injection tricked it into exposing its system prompt. 3️⃣ In the process-guided agent, the same attacks failed. Not because of clever prompt engineering, but because user input never routes directly to instruction interpretation. The tradeoff: → Prompt-driven = flexibility + wide attack surface → Process-guided = fewer paths + minimal exposure The report details real exploits and why some architectures pose risk while others contain it. If you’re building agents for production, check out this report to learn more. 👉 Full Lakera assessment: https://hubs.ly/Q03P9XKm0 🔧 Try Rasa: https://lnkd.in/dXFrPWH9
-
-
Building a reliable conversational AI assistant demands clear control, transparency, and real-world validation. Our readiness assessment shows teams exactly what’s required to scale confidently from pilots to full production. Spend 5 minutes to discover your AI’s strengths and gaps with a practical roadmap for growth. 👉 Start the assessment now: https://hubs.ly/Q03Cj_HN0 #ConversationalAI #GenerativeAI #Enterprise #Automation #AI #Rasa
-
-
"We needed Grace to understand when a patient says 'I need to change my appointment' while they're in the middle of asking about lab results." Providence Health’s engineering challenge: Build an AI that stays conversational but never loses track of what the patient needs to accomplish. Their solution separates the "understanding" from the "doing". LLMs handle conversation nuance while structured flows reliably execute healthcare workflows. Providence's cross-functional team built Grace to handle the complex realities of healthcare interactions, where patients routinely switch between different needs within a single conversation. Grace now handles 160K+ conversations monthly with 59% task completion. 📊 👉 Read the full story: https://hubs.ly/Q03M52Hx0 #Healthcare #ConversationalAI #ProductDevelopment #PatientFirst #HealthTech #Rasa
-
-
How secure are different AI agent designs? We partnered with Lakera, an independent AI security firm. The research, conducted independently by Lakera, tested two agents under the same conditions: 1️⃣ A traditional prompt-driven agent 2️⃣ A Rasa process-guided agent Both had the same functionality. Both faced the same real-world attack attempts: 🔹 Prompt injections 🔹 Data extraction 🔹 Jailbreaks 🔹 Content safety violations 🔹 Infrastructure attacks The outcome: ❌ The prompt-driven agent showed critical vulnerabilities in every category ✅ The Rasa agent contained the attacks, limiting the impact to minor variations The key insight: It’s not about which LLM you use. 👉 It’s about the architecture of the agent. Prompt-driven designs offer flexibility, but also introduce risks that are challenging to manage at enterprise scale. 🔗 Read Lakera’s full security assessment: https://hubs.ly/Q03N4ggM0 🔗 Try Rasa to see for yourself: https://lnkd.in/dXFrPWH9
-
-
"We need to understand exactly what this AI agent can and cannot do before approving production deployment." Every enterprise security team asks this question. Most AI systems can't provide a satisfying answer because their architecture doesn't support that level of control. Our white paper shows how CALM enables complete traceability from user input to system action, with policy enforcement through code rather than prompts. When security teams can see exactly how agents behave under any condition, AI transforms from compliance risk to competitive advantage. 👉 Get the white paper: https://hubs.ly/Q03Ll0mN0 #Enterprise #Security #Compliance #ConversationalAI #Developers #Rasa
-
-
That's a wrap on Customer Engagement Summit 2025! Thank you to everyone who stopped by our booth and attended James Mackay's session on The Future of the Contact Centre stage. The energy and conversations around transforming customer engagement were incredible. It's clear that organizations are ready to move beyond basic chatbots to AI agents that truly understand their customers.
-
-
A leading insurer needed AI assistants that could scale across service, sales, and voice. By consolidating logic with Rasa CALM, they simplified development and supported over 1 million voice conversations annually. See how they increased deflection, reduced fallbacks, and gained full control over system behavior. 👉 Read the case study: https://hubs.ly/Q03MVdd20 #ConversationalAI #Insurance #VoiceAI #Agents #CustomerService #Rasa
-
-
Your automation reduces agent workload by 40%, but your budget still gets questioned during planning cycles. Reporting "improved response times" instead of "reduced operational costs" makes even successful AI look like overhead to finance teams who evaluate investments based on financial impact. Executive stakeholders need cost-per-resolution data, not accuracy improvements. 👉 ROI measurement framework: https://hubs.ly/Q03Lw9cG0 #Enterprise #ConversationalAI #GenAI #ROI #Budget #Operations #LLM #Rasa
-
-
We're excited to announce that Rasa is a part of the Service Summit in Hamburg this November! Find the Rasa team at our booth to see how we’re redesigning the customer experience. Book a meeting with us on November 19-20 here: https://hubs.ly/Q03LVF3p0 #ServiceSummit2025 #CustomerExperience #ConversationalAI #Hamburg
-