We’d love for you to join Basis.
Rapid advances in frontier AI are shaping how billions of people work, learn, & live every day. Much of this advancement is due to scaling (more compute + more/better data → better coverage over desired distributions). But, while compute spend is growing up to 4x per year, publicly available data increases by only 1.03x per year! That gap becomes even starker when moving beyond LLMs into voice, video, & the physical world. Hence, there is desperate demand for high quality, high diversity, & carefully curated recordings of human interaction and expression.
Leading labs require vast amounts of human feedback, behavior, culture, speech, & judgement. We’re already working with leading labs and have one of the largest contributor networks in the industry, to provide exactly that.
Our mission is to revolutionize the frontier of human-AI interaction through consumer products users genuinely want to use. The intelligence we provide today will power the world’s best models to better serve humanity, its interests, & its values in the long-run.
Our mission works two-ways:
We give frontier labs access to an increasing breadth of human intelligence and experience, helping them build models that are more capable, grounded, and useful to the people they will serve.
We give humans everywhere a meaningful role in shaping those systems and reward them for the value they create.
Why Basis will win / what we do:
This sector will only continue to grow. Yet almost all companies serving this market only operate as marketplace middlemen and temporary talent acquisition portals. Labs needs data, the vendor recruits costly contractors, & the workforce largely gets reset when the project ends. Advantages are slow to compound and there is little to nothing proprietary to keep them afloat once the sector matures and becomes fully efficient.
We are not building middleman infra between labs and contractors. We build consumer-facing products millions genuinely want to use, reward them for participating, and own the infrastructure that turns user behavior into high-quality training signals for labs, with compounding network effects, high k-factor and retention, and real-time rewards aligned with contribution quality.
We are experts in frontier multi-modal research. We collect tens of thousands of hours of recordings and millions of preference votes weekly (and growing rapidly), and have already used these to train SOTA internal models to optimize all parts of the pipelines. We allocate significant GPU resources to make the best use of our proprietary data and constantly test new training hypotheses. We will open-source and publish portions of this work to support the community and show the breadth of our expertise.
We are experts in reaching and retaining consumers. We have previously scaled consumer apps from zero to millions in ARR and have backgrounds at the best consumer companies in the world (Duolingo, TikTok, Zynga, etc.).
Our goal is to build the world’s largest consumer products for teaching AI about humanity and the real world: a connection through which frontier models can learn from millions and soon hundreds of millions of people every day. Very few companies are attempting to build this system and even fewer have the skills to actually do it. We’re confident Basis will become the defining company in this category.
Why you shouldn’t join:
Ambiguity: We will spend considerable time teaching you everything we know about this field, our product, our needs, & our systems. Once we’ve shared the context, you’re expected to make independent decisions and lead the way through the ambiguity that comes with working on the frontier. You will become the world’s leading expert on whatever you work on here, and won’t settle for less.
Fast Pace: We don’t expect you to work outside of our work hours (weekdays 10-8), but we do expect you to move fast, parallelize relentlessly, & get client projects across the finish line. We are competing in what is turning out to be one of the largest emerging markets and we plan to win it.
Small Team: We’re a small team (4 full-time, 2 part-time) & hiring 2-3 in the next few weeks. If you prefer the comfortable structure of a larger company don’t join us yet!
Evolving role: This is a startup & we’re scaling aggressively - it’s insane how much can change within just one year. We expect that everyone’s role will also evolve aggressively; you are expected to grow with the company and will need to constantly learn new things - every day brings new challenges & no two days are the same.
Why you should join:
You’ll work on the cutting edge of frontier AI working directly with top labs. You’ll build systems/train models/do things no one has ever done before.
You’ll work with exceptional, ambitious, & fast-moving people while having full ownership & creative control over your work.
You’ll grow & learn a lot. Your career will accelerate with the company. Basis is well-funded from S-Tier investors and inbound, and will imminently raise a Series A (now’s the time to join!).
What you’ll do (Software Engineer):
Build pipelines that score, filter, and rank millions of user recordings each day using in-house ML-models and custom fraud detection systems
Design scalable systems from scratch that automatically control millions of dollars in user rewards and ad spend.
Work directly with leading labs, assess their needs, & deliver high-quality and heavily filtered recordings for their frontier models.
Collaborate closely with founders in a fast-iteration environment, independently owning systems end-to-end with high autonomy.
You: Enjoys system design; is very intentional & sweats the details; moves very fast with AI coding tools; loves building; thrives on solving problems no one has ever solved before.
What you’ll do (Research Engineer):
Train state-of-the-art internal models, public benchmarks, and technical publications using the one-of-a-kind multi-modal data we collect.
Collaborate with top researchers at leading labs to create extremely challenging open-source benchmarks to push the frontier of voice and interaction modeling.
Example: frontier speech-to-text models still struggle with regional dialects like Moroccan Arabic. We collect tens of thousands of hours of hand-transcribed Moroccan Arabic audio and use it to train internal models that accelerate transcription and automatically verify data quality.
Example: we collect tens of millions of preference ratings to train in-house reward models used in RL pipelines, for both enterprise clients and open-source releases/benchmarking.
You: deep conceptual understanding of deep learning; experience fine-tuning strong open models; ideally with publication track record.
What you’ll do (MTS/Research-Software Hybrid Role):
Both of the above
Employment Details:
💸 Competitive compensation + equity
🏝 20 days PTO
🌺 Company retreat
🏥 Platinum-class medical, dental & vision insurance
🍱 Lunch & dinner provided every work day (weekdays, 10-8)
🏢 Beautiful office space in San Francisco
🌎 Best-in-class immigration support
❤️ Most importantly, the ability to work on some of the most consequential technical challenges of our time.
We'd love to hear from you. Reach out to alexis@withbasis.co with any questions!