The last two years at Turso have seen me wear many hats. Developer Advocate, Engineer, Marketer, Writer, and Product tinkerer.
I’ve made the decision to take on a new challenge, and will be leaving Turso at the end of October, marking my two year anniversary. Here’s a look back at what I achieved.
I joined Turso in late 2023, as part of a small but ambitious marketing team. The company was mid-rebrand, introducing a new identity around bringing SQLite to production. At the time, “SQLite in production” wasn’t really a thing. Turso was one of the first to seriously push SQLite as a modern database for distributed, edge-first applications.
Goal: Make Turso easy to understand, adopt, and integrate into any workflow.
From day one, it was clear the documentation wasn’t keeping up with the product. So I rebuilt it from the ground up, restructured around API, SDKs, and Dashboard. I tested every endpoint, reported bugs, and even fixed a few myself.
As the product matured, I expanded the docs to cover emerging patterns like multi-tenancy (“database per user”) and vector embeddings. I wrote hands-on guides for deploying embedded replicas on platforms like Railway, Fly.io, and Render, showing developers that Turso could fit into any workflow.
When the Dashboard needed a refresh, I rebuilt it during the team’s Brazil offsite. The old version was functional but bolted together. The new version (launched in August 2024) was decoupled, cohesive, and dramatically improved how we could ship product updates.
Later, when Turso migrated to AWS and deprecated Edge Replicas, I built a DIY CDN proof of concept using Fly.io’s persistent storage and anycast network. It gave developers a path to edge replication even without official support.
The docs got a complete overhaul, the dashboard finally felt right, and developers had guides for pretty much any platform they wanted to use.
Goal: Help developers understand Turso’s capabilities and see what’s possible.
I wrote dozens of blog posts and recorded videos explaining core concepts. Early on, I published Give Each of Your Users Their Own SQLite Database, which became one of our most-read posts and helped position Turso’s multi-tenancy story. By May 2025, Turso crossed 2 million databases, validating that we were onto something with small ephemeral databases for the multi-tenant use case we’d been pushing so hard on.
To grow adoption of the Platform API, I created a six-part blog and YouTube series (“A Hero’s API Journey Begins”) showing how Turso wasn’t just a hosted database, it was a platform for platforms.
When Turso launched vector support, I built a LangChain integration for libSQL and wrote tutorials bridging SQLite, AI, and embeddings:
I also demonstrated multi-tenancy patterns in posts like Working with Clerk and Per-User Databases, which is also available as a Vercel Template.
Later, I prototyped the Turso Schema Registry during an Istanbul offsite with Pekka. That late-night hackathon turned into a working solution for deploying schema changes across millions of databases.
Between the blog posts, video tutorials, LangChain integration, and proof-of-concept tools, we had a ton of educational content showing what Turso could do.
Goal: Build Turso’s presence and make it feel everywhere.
In March 2024, we ran a Launch Week. In the weeks leading up, I rebuilt the entire marketing website. It gave the brand a fresh look to match the rebrand, and we launched it alongside Launch Week.
That week marked the moment Turso started to feel everywhere.
For our second Launch Week in October 2024 (my one-year anniversary), I went all-in on a Halloween theme, inspired by my daughter sticking a Turso sticker on a pumpkin. I designed spooky emojis for Discord, Halloween-themed blog covers, and updated the website with a dark theme (bats, pumpkins, the works) for Launch Week posts.
In August 2025, I led a migration away from Customer.io and HubSpot to Resend for both newsletters and transactional emails. Our existing email templates hadn’t been updated in years and were looking dated and inconsistent with our brand. I consolidated our email infrastructure into a single platform, completely refreshed the template design with a modern look and feel, and restructured the Go backend to handle event tracking for customer usage notifications, blocked access alerts, and payment confirmations.
Two successful Launch Weeks, a rebuilt marketing site, modernised email communications, and some genuinely fun brand moments that people remembered.
Goal: Make Turso work seamlessly with the tools developers already use.
As the user base grew, we noticed developers from Rails, PHP, Java, and React Native communities showing up in Discord, but our SDKs were focused on Node, Go, and Rust. We hired someone full-time to manage SDKs for the core database, and I partnered with them to build bridges between those SDKs and ecosystem tools. I helped expand ecosystem support by building SDKs for Shopify (with a Drizzle contributor), Sentry, and GitHub Actions.
In mid-2025, I rebuilt the Vercel Marketplace integration from scratch in Go (despite barely knowing Go). Nikita’s PR reviews and attention to detail helped me grow my confidence in Go along the way. The integration let developers create and manage Turso databases directly from Vercel’s dashboard, billed through Vercel. It launched in June and is now installed hundreds of times every week.
We ended up with SDKs and integrations across a bunch of platforms (Vercel, Shopify, Sentry, GitHub Actions), which opened Turso up to way more developer communities.
Goal: Build a thriving, engaged community around Turso.
By early 2025, our Discord had grown to over 6,000 members. Everyone at Turso is active there, especially during on-call rotations. The vibe has always been relaxed and friendly, less like support and more like a group of people building cool things together.
To make it easier to recognize paying customers and their tiers, I built a Discord verification bot that gave users badges for their plan and changed their username color based on their tier. It helped the team quickly identify who was on which plan when jumping into support threads.
To keep that energy going, I kicked off monthly Community Calls (first Thursday of every month). Glauber and I would spend 30 minutes doing Q&A, sharing product updates, and talking openly about what’s next. Having direct access to the CEO was something users consistently appreciated.
Every Wednesday, I’d post a “What’s Happening This Week” message in Discord, sharing new blog posts, community projects, and upcoming events. It became a rhythm that kept everyone in the loop and turned Discord into a living part of Turso’s community.
We grew the Discord to 6,000+ members, ran monthly calls with the CEO, built tools to better support paying customers, and created a community that actually felt like part of the product journey.
Goal: Support the launch of Turso’s next-generation SQLite database.
Over the last few months, Turso was gearing up for its biggest moment yet: the Beta Week launch of the Turso Database rewrite. This was the next evolution of SQLite, and I worked with the team to bring it all together.
I collaborated on blog posts, refreshed the marketing website, and created content for SF Tech Week. Working with Nikita on Turso Sync and browser examples, I built the Turso Shell, an in-browser tool that let developers demo what we’d built and get started with the new database instantly. It became an easy way for people to experience the power of Turso without any setup — even Vercel CEO Guillermo Rauch posted about it on X!
Beta Week was the culmination of everything we’d been building toward, and it felt like the right note to end on.
Two years, dozens of blog posts, hundreds of docs pages, thousands of Discord messages, and a ton of learning.
Working at Turso has been a rare experience: building something genuinely new, alongside some of the best engineers I’ve ever worked with.
I’d like to thank Glauber, Pekka, and the whole team for the opportunity to work on an ambitious project. I’ll be advocating for Turso from the sidelines.