Technical Growth Lead — ML Community & Developer Advocacy
Be the first technical face ML engineers meet when they discover tracebloc — community, content, conferences, and credible developer advocacy.
What you'll do
You'll be the person ML engineers and data scientists meet first when they discover tracebloc. Your job is to make that discovery happen — in Reddit threads, on HuggingFace, in Discord conversations, at conferences, and through the technical content you create. You're not a marketer who learned to talk to developers. You're a developer who figured out that helping people find the right tools is more fun than building pipelines. Day to day, you'll hunt for conversations where someone is stuck on a problem tracebloc solves — a team struggling to benchmark models across sites, a researcher who can't get access to private data for training, a data scientist trying to compare vendor models without moving data. You'll jump in with a genuine, technically credible response. Sometimes that means writing code. Sometimes it means linking to a Colab notebook you built. Sometimes it means just being helpful. The tracebloc workspace link comes naturally when it's actually the right answer. You'll also build and run our developer community from the ground up, create technical content that ML practitioners respect (blog posts with code, not whitepapers with buzzwords), speak at conferences alongside our CEO, and join enterprise calls when prospects need someone who can explain how federated learning actually works under the hood. This is an early-stage role. There is no playbook waiting for you (well, there is a messaging playbook — you'll use it). You'll propose hypotheses, run experiments, measure what works, and iterate. If Reddit isn't working this week, you'll try something else. You're a generalist who happens to be deep in ML.
What we're looking for
- 2–4 years of hands-on ML or data science work — you've trained models, written PyTorch or TensorFlow code, used tools like HuggingFace, W&B, or Kaggle
- Visible activity in ML communities — a post history on Reddit, HuggingFace contributions, Kaggle notebooks, technical threads on X, a blog, a YouTube channel, or some combination
- You can write technical content with code snippets that ML engineers actually read and learn from
- Comfortable presenting at meetups and conferences — energetic, credible, not corporate
- You think in experiments: "I'd try X because Y, and I'd measure it by Z"
- You don't need a manager to tell you what to do on Monday morning
Nice to have
- Experience managing a Discord or Slack community
- Understanding of federated learning or privacy-preserving ML
- German language skills
- Experience with Kubernetes, Docker, or ML infrastructure
- A personal following in the ML space (any size)
What we offer
- Work on a product that solves a real problem for ML teams — collaborative AI development without moving data
- Optional equity participation in a company building a category nobody else owns
- Direct work with the founding team — your ideas ship, your experiments run, your voice matters
- Berlin office
- Conference travel budget — you'll be representing tracebloc at events
- The chance to build a developer community from scratch with real resources behind it
How to apply
- Don't send a cover letter. Instead, find a thread on Reddit or HuggingFace where someone has a problem tracebloc could help with. Write the reply you'd post.
- Send that, along with your favorite piece of technical content you've created (blog post, notebook, tutorial, thread — anything), to careers@tracebloc.io.