採用
必須スキル
Machine Learning
About Comet
Comet is building the development platform for teams who want to ship robust, reliable, and responsible AI applications. Opik, our open source LLM evaluation framework, has quickly become one of the most popular tools in the space. Our experiment management platform is used by data scientists at companies like Uber, Netflix, and Etsy. Tens of thousands of researchers, engineers, academics, and hobbyists use Comet every day to build the future of AI.
Working at Comet will give you access to the most exciting work being done in all areas of machine learning. Some of the top researchers and companies working on self-driving cars, drug discovery, particle research, diffusion models, and LLMs use Comet every day. Your work has the potential to accelerate the development of some of the most impactful technology in the world, and you will be doing it alongside a team of passionate, caring individuals. If that sounds exciting, Comet is the right place for you.
Comet is backed by more than $63 million in venture capital funding and powers some of the best machine-learning teams in the world, including Netflix, Uber, Etsy, and Mobileye. We are a remote-first company with offices in New York City (USA) and Tel Aviv (Israel).
The Role
We're looking for a Dev Rel who lives and breathes the AI builder ecosystem: someone who makes great content, gets excited about new models and frameworks the day they drop, and knows how to communicate technical ideas in a way that makes developers stop scrolling.
You'll be an IC embedded in the Dev Rel function, working closely with the broader marketing team. Your north star is developer trust and adoption, earned through authenticity, technical credibility, and genuinely useful content, not marketing fluff.
If you've been building a following by showing people how to actually use agents in production, and you want to do that full-time with a product that developers already love, this is your role.
What You'll Do
Content Creation (your primary output)
-
Produce high-quality technical video content: tutorials, demos, explainers, live coding sessions: the kind of content the AI community actually watches and shares
-
Write technical blog posts, notebooks, and guides that showcase Opik in real-world agentic workflows
-
Build creative technical demos that highlight use cases for evaluation, tracing, and observability
-
Show up on social (X, LinkedIn, YouTube) as a genuine voice in the AI developer ecosystem
Community & Ecosystem
-
Be active in the communities where our users live: Discord, Slack groups, Reddit, GitHub, etc. discussions
-
Attend and speak at SF-based and national AI events (conferences, meetups, hackathons)
-
Identify and engage with influential developers and open source contributors
-
Run or support technical workshops, paper readings, and community programming
Product Feedback Loop
-
Surface developer sentiment and friction points back to Product and Engineering
-
Be a voice of the community inside the company: what are developers actually struggling with?
What You Bring
-
3+ years of experience in developer advocacy, developer education, or a closely adjacent technical role
-
You are genuinely plugged into the AI ecosystem: you know the key researchers, practitioners, open source projects, and communities
-
Strong video content chops: you can script, record, edit, and publish compelling technical video without a production team
-
Technical depth: comfortable coding in Python, building LLM applications, and talking confidently about evals, agents, and observability
-
A natural communicator who can simplify complex ideas without dumbing them down
-
Preferably based in San Francisco or the Bay Area: you'll be embedded in the local AI scene and representing Opik at events
Nice to Have
-
An existing following or presence in the AI/developer community (YouTube, X, newsletter, podcast)
-
Experience with LLM evaluation frameworks, tracing tools, or observability platforms
-
Open source contributions or maintainer experience
-
Startup experience: you've worked in a fast, scrappy environment before
Why Comet
-
Your work will directly drive adoption of a product developers already love
-
You'll have creative autonomy: we want your authentic voice, not a brand script
-
SF-based role with a remote-first team
What We Offer
-
Competitive Base salary - $180-250K, based on proven experience, skills and location.
-
Competitive benefits package.
-
Flexible working hours and remote work options.
-
Opportunities for professional growth and development.
-
A collaborative and innovative work environment.
-
The chance to work with cutting-edge technologies and projects.
-
This role will be located in the USA, Preferably based in San Francisco or the Bay Area, working with a global team (large presence also in Europe and Israel) - some flexibility with work hours is required.
Comet is an equal-opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees without regard to race, religion, color, sex, gender identity, gender expression, sexual orientation, national origin, ancestry, citizenship status, uniform service member status, marital status, pregnancy, age, medical condition, physical or mental disability, genetic information/characteristics, and any other characteristic protected by State or Federal law.
総閲覧数
0
応募クリック数
0
模擬応募者数
0
スクラップ
0
類似の求人

Developer Relations Engineer, Cloud Platform Evaluations Team
Google · placeKirkland, WA, USA; Sunnyvale, CA, USA; +2 more; +1 more

CONTRACT - Marketing Coordinator, Publisher Developer Relations
Nintendo of America · Redmond, WA

Developer Relations
Railway · Remote (United States)

Developer Advocate
Figma · San Francisco, CA

Developer Relations Engineer
Sierra AI · New York, NY
Comet MLについて

Comet ML
Series BComet ML provides a machine learning experiment management platform that helps data scientists and ML engineers track, compare, and optimize machine learning models and experiments.
51-200
従業員数
New York
本社所在地
レビュー
3.7
1件のレビュー
ワークライフバランス
3.0
報酬
3.0
企業文化
3.0
キャリア
3.0
経営陣
3.0
80%
友人に勧める
良い点
Enables visualization and automatic tracking of dataset work
Useful integration for EDA workflows
Good technical functionality
改善点
Limited feedback available
Single review basis
No specific cons mentioned
給与レンジ
5件のデータ
VP
VP · Chief Executive Officer
2件のレポート
$172,500
年収総額
基本給
$150,000
ストック
-
ボーナス
-
$172,500
$172,500
面接体験
50件の面接
難易度
3.2
/ 5
期間
14-28週間
内定率
38%
体験
ポジティブ 65%
普通 18%
ネガティブ 17%
面接プロセス
1
Phone Screen
2
Technical Interview
3
Hiring Manager
4
Team Fit
よくある質問
Technical skills
Past experience
Team collaboration
Problem solving
ニュース&話題
is it safe to click a promoted post here on reddit? (website version)
i accidentally clicked on a promoted post here on reddit (the one that says Opik) and it redirected me to a different site which i immediately exited. can that give me a virus? or any spyware/malware? is that a malicious link/post? the screenshot is from my other post that i posted on a different community since i already deleted the original screenshot.
·
4w ago
·
1
·
1
clicked on a promoted post here on reddit
i accidentally clicked on a promoted post here on reddit and it redirected me to a different site which i immediately exited. (the one that says Opik), can that give me a virus? is that a malicious link/post?
·
4w ago
·
6
Hugging face is somewhere to share open LLM models, not actual faces.
Poor girl posts a selfie to the hugging face subreddit which is a sub for talking about [huggingface.co](http://huggingface.co) \- a website for sharing open source LLM models.
·
4w ago
·
35
·
3
5 OpenClaw plugins that actually make it production-ready
If you are using OpenClaw regularly, you start noticing patterns after a while. Simple tasks hit the same expensive model used for complex reasoning. MEMORY. md keeps growing but still misses the right context. Connecting tools like Slack, GitHub, or Gmail means dealing with auth and tokens. Sometimes the agent just stops mid-run and you have no clear trace of what happened. The agent works, but the system around it feels inefficient. Most of this comes from how the default setup works. It re
·
4w ago
·
191
·
72