The posting
ABOUT US
We're building toward a world where every company can become its own AI lab.
Goaly is a stealth AI startup founded by ex-Meta Superintelligence Labs engineers and researchers. Our mission is to dramatically lower the cost, time, and talent barriers to building proprietary AI — and make each generation of models faster and cheaper to build than the last.
Backed by leading AI investors and endorsed by frontier AI researchers and builders, we're looking for an engineer who can build in public, explain hard systems clearly, and already knows how to reach the people who train models.
ABOUT THE ROLE
Goaly's users are the people who run post-training: the ML infra engineers and researchers who babysit SFT, RL, and fine-tuning jobs on GPU clusters. They don't read marketing. They read benchmarks, repos, docs, and the engineer who wrote them. You will be that engineer.
You will be our first Developer Relations hire and the technical voice of Goaly as we come out of stealth. Half of your week is building: running real post-training workloads on our stack, writing the benchmark harnesses, example integrations, and docs that get a stranger from zero to a first result in thirty minutes. The other half is making sure the right ten thousand people see it: engineering posts, talks, open-source releases, and the one-on-one conversations with practitioners that turn into design partners and hires.
This is not a marketing role with a technical veneer, and it is not an engineering role that occasionally posts. You need to be credible in a room full of people who train models, and you need an audience in that room — or a proven ability to build one.
This is a founding role. You will report to the founders and work daily with engineering and the go-to-market team. There is no playbook yet; you help write it.
WHAT YOU'LL DO
- Build the signature technical artifacts. Own a public post-training efficiency benchmark, reproducible before/after case studies (doomed-run kills, silent throughput regressions, dollars per eval point), and the open-source tools we release around them.
- Write the docs that convert. Quickstarts, reference docs, and example integrations for the stacks our users actually run (PyTorch distributed, Megatron, FSDP, TorchTitan, Ray, Slurm, Kubernetes on GPUs, W&B, MLflow) so an engineer can get a read-only Retroactive Kill Report on their own cluster without talking to us.
- Create technical content practitioners share. Deep-dive engineering posts, benchmark write-ups, failure-mode teardowns ("seven ways an RL run dies silently"), short demo videos, and live walkthroughs.
- Own distribution. Run Goaly's technical presence on X, LinkedIn, GitHub, Hacker News, and the ML-infra Discord and Slack communities. Land talks at the venues our users attend (PyTorch Conference, MLSys, Ray Summit, GTC, NeurIPS and ICML workshops, Bay Area AI infra meetups). Build relationships with the people whose posts our audience already reads.
- Be the feedback loop. Sit with design partners, reproduce their problems on our stack, and bring engineering precise, prioritized feedback. Turn repeated questions into docs, tools, or product changes.
- Build community around the problem, not the product. Practitioner conversations, office hours, and a newsletter or series that becomes the place to learn about post-training efficiency.
- Amplify the founders. Turn the founding team's Llama post-training infra experience into content and talks, and work with the GTM team to turn attention into design-partner pipeline and candidate pipeline.
- Measure and cut. Track reach, repo adoption, docs funnel, inbound design-partner requests, and hires sourced. Double down on what works and kill what doesn't.
YOU MAY BE A GOOD FIT IF YOU HAVE
- A real engineering background. 3+ years as a software or ML engineer, and you still write code every week. You can stand up a distributed training job, read a profiler trace, and debug a CUDA OOM without asking for help.
- Hands-on time with LLM post-training or the infra under it. SFT, RL, LoRA, DPO, or the stack beneath them (PyTorch distributed, Megatron, DeepSpeed, FSDP, Ray, Slurm, GPU Kubernetes). You don't need frontier scale, but you need to have felt the pain.
- A track record of explaining hard things clearly. Blog posts, docs, conference talks, OSS READMEs, tutorials, or videos. Send links; we will read them.
- Distribution you can point to. An audience on X, LinkedIn, YouTube, a newsletter, or GitHub; talks you've given; communities where people know your name; or a demonstrable history of building reach from zero. Reach inside the ML infra and post-training community matters far more than raw follower count.
- Communication in both directions. Writing that practitioners trust, and the ability to listen to a frustrated infra engineer for an hour and come back with the three things engineering should fix.
- Taste for substance. You would rather publish one reproducible benchmark than ten hot takes.
- Comfort being the public face of something early. No playbook, an imperfect product, shifting messaging, and your name on the output.
STRONG PLUSES
- Prior developer relations, developer advocate, or technical evangelist experience at an ML tooling, AI infra, or developer-platform company (Weights & Biases, Modal, Anyscale, Together, Hugging Face, Lightning, Databricks, or similar).
- Maintainer or notable contributor to a widely used ML or infra open-source project.
- Experience running a developer community: Discord or Slack, meetups, hackathons, workshops.
- Fluency on video and on stage; comfortable on a podcast or a conference keynote.
- Daily familiarity with the observability and profiling tools our users live in (W&B, MLflow, Nsight, PyTorch Profiler).
- Sales-engineering or forward-deployed engineering experience: you have gotten a customer to a result, not just watched them try.
HOW WE WORK
- Mission first. We choose work for its impact on the mission and take responsibility for the outcome, not just our assigned tasks.
- High agency. We identify what is missing, form a plan, and move without waiting for perfect clarity.
- Speed with rigor. We ship, measure, and iterate quickly while protecting correctness, safety, and reliability.
- Flexible scope. We cross team and technical boundaries when that is the fastest way to solve the real problem.
- Low ego, high standards. We give direct feedback, change our minds when the evidence changes, and help the whole team win.
- Continuous learning. The stack changes quickly; we are willing to learn unfamiliar systems, methods, and domains as the work demands.
LOCATION, VISA SPONSORSHIP & BENEFITS
- Hybrid in Palo Alto: 4+ days/week in office, with roughly 20% travel for conferences and customer visits.
- Visa sponsorship: H-1B and OPT/CPT support available, with immigration counsel.
- Meals & perks: Complimentary lunch, dinner, snacks, and drinks.
A note on qualifications. We value exceptional ability over perfect keyword matches. If the work excites you and you can show strong technical ability, clear writing, and real reach, we encourage you to apply.
EQUAL OPPORTUNITY
We are an equal opportunity employer. We consider qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity or expression, national origin, age, disability, veteran status, genetic information, or any other characteristic protected by applicable law. We provide reasonable accommodations for candidates who need them during the hiring process.
