GopherCon 2025 was an incredible week. We shared Diagrid’s Workflow Composer on the main stage to nearly 1,000 attendees, hosted several community meetups, and connected with industry leaders to discuss Dapr Workflows and Catalyst. With strong themes around AI and observability, the conference was a huge opportunity to showcase how Dapr + Diagrid stand out as a powerful, self-hosted and enterprise-grade workflow solution with AI capabilities.
Last week Cassie and I had the privilege of attending (and speaking at!) GopherCon 2025, and what a week it was. With nearly 1,000 attendees, single-track programming, and strong themes around AI, observability, and tracing, it was an incredible chance to connect with the Go community, share what we’ve been working on, and learn where the industry is heading.
Highlights from the Conference
Mainstage Talk
I had the chance to present our Workflow Composer to a packed room of 800–1,000 people. It was energizing to highlight the power of Dapr’s Workflow API and how Diagrid makes it easy to create and manage Dapr Workflows. The turnout and follow-up conversations showed just how much interest there is in practical cloud-native workflow solutions in the Go community.
Meetups for Community
Cassie and I also hosted the Women Who Go event and a Meetup for Meetup Organizers, each drawing 50+ attendees. These were vibrant, community-first conversations where we had good conversations around Go and the evolving AI ecosystem, and also discussed Dapr and Diagrid. We met developers interested in applying for our open roles, which was super exciting. Building community is a cornerstone of what makes events like GopherCon so valuable, and these sessions reminded us how much energy there is when developers get together in person.
Conversations with Industry Leaders
One of the most exciting connections we made was with the author of the excellent open-source durable execution repo where he expressed interest in Diagrid adding further Dapr content for reference for his community. We also met with financial leaders, discussing workflow engine needs and performance capabilities, as well as another company excited to try Dapr in an upcoming project of theirs! Many attendees were hearing about Dapr for the first time and were excited to learn that it supports workflows, in addition to all of the other enterprise-grade APIs. A frequent reaction: “I didn’t know Dapr could do this!” with a smile and nod of approval!
What We Learned
AI Everywhere
Many talks explored practical challenges in AI systems—from LLM reliability (routing between providers like Anthropic and OpenAI) to performance tuning, chatbot latency, and multi-agent system design. We drew parallels to the familiar challenges of microservices and distributed systems, now resurfacing in the context of agentic systems.
- One AI talk on Scaling LLMs with Go emphasized the fundamentals of chatbot design: minimizing latency, improving voice realism, and refining turn detection.
- Agentic systems are reinforcing what many of us have seen in distributed systems: multi-agent approaches outperform monolithic agents, though the learning curve is steep and best practices are still emerging.
- The co-creator of MCP at Anthropic noted a growing focus on agentic CLIs, integration points with MCP, and tracking where investment dollars are flowing in the agent ecosystem.
Go Innovations
- Green Tea garbage collection will become the default in the next Go release, promising lower CPU costs and better performance.
- New FIPS 140-3 compliant post-quantum secure connections in crypto/tls are a big milestone for Go’s security.
- Profiling overhead remains below 2%, meaning production teams can enable tracing/profiling with confidence which is both exciting and useful.
- Go now has a dedicated MCP package with an entire team supporting it. A full talk at the conference dove into how MCP is shaping integration and collaboration patterns in Go ecosystems.
Reminders
- Pin by commit, not by tag. Git tags are mutable, which can lead to subtle dependency issues. The community highlighted tools like Go Capslock for capability checks and code analysis to help mitigate this.
Opportunities for Next Year
Diagrid’s focus remains on community presence, which is core to our industry, products, and culture. We’ll continue supporting diversity and inclusion through events like Women Who Go and the Meetup for Meetup Organizers, helping foster connections across the Go community. And of course, we’ll do our best to be back on stage at GopherCon, showcasing the innovative and fun projects we’re building at Diagrid!
Closing Thoughts
GopherCon 2025 was a huge success for us. From sharing Dapr workflows and highlighting Diagrid’s Workflow Composer on the mainstage, to connecting with community leaders, to learning where Go is headed, the conference reaffirmed how much opportunity there is to highlight what Dapr and Diagrid are building and where we’re headed next.
If you haven’t yet explored our workflow offerings, then keep an eye on the updates we’ll be rolling out in the coming weeks. We’ll be sure to share the link to the recording of Sam’s talk at GopherCon this year! We’re excited to keep pushing forward and to see even more Gophers adopting Dapr in 2025.
- Curious about Dapr on Go? Check out the Go SDK here.
- Jump into the Dapr Quickstarts to explore Dapr’s many APIs, each with examples in Go.
