Website:
lh2.ai
Job details:
Mandatory requirement: Graduate from a Tier 1 engineering institution such as IIT, BITS, NIT, IIIT, or equivalent.
About LH2 AI Labs
Built by second-time founders, LH2 AI Labs is building the post-training infrastructure for frontier AI models. We bring private, institutional datasets and vetted domain experts into frontier AI pipelines across verticals such as coding, computer use, agentic workflows, medical and audio.
The public web has been crawled and trained on, and there is little new signal left there. We work with data that has never left the companies that produced it. Our vision is a world where frontier models get high-quality data on tap, the way they get compute today.
Mission
Own LH2's data marketplace as a product, on both sides. For companies, that's discovering what their data is worth and sharing it without anything sensitive leaving their environment. For frontier labs, it's finding, evaluating and licensing data they can't get anywhere else.
Problems you'll work on
- What does a company need to see before it trusts us with years of its internal data?
- How do you price an asset nobody has priced before, from a small sample?
- What makes a dataset worth buying to a frontier lab, and how do we know before we build it?
- What is "clean enough," and how do you make it a promise both sides will sign?
- Which parts of this marketplace should be software, and which should stay human?
What you'll own
- The marketplace roadmap, from raw institutional data to a delivered, licensed dataset.
- The experience on both sides: data owners, and the labs that buy.
- The bar for "sellable": packaging, format, provenance and rights standards.
- Provenance, rights clearance and privacy as product features, not compliance checkboxes. They're the trust the business rests on.
Where we are
Early. The founding engineering team is being hired now, and the first product spec for the marketplace doesn't exist yet. You'll write it.
Must-have
- 5+ years in product, with real zero-to-one ownership. You've defined and shipped a technical product, not just managed a backlog.
- Comfort with technical, data-heavy or ML-adjacent products. You can hold your own with researchers and engineers.
- Strong customer discovery with sophisticated technical buyers.
- The judgment to say no and keep a small team focused.
- Clear written thinking that a lab and an engineer both understand.
- A degree from a Tier 1 engineering institution (IIT, BITS, NIT, IIIT or equivalent).
Nice to have
- Experience in AI/ML data, developer tools, data infrastructure or B2B data products.
- Understanding of post-training (SFT, RLHF, evals, RL environments) and what frontier labs buy.
- Products where privacy, security or compliance shaped the design.
- Two-sided marketplaces, or products where pricing was the product.
Click on Apply to know more.
