Example Company
Software engineer
- Location
- San Francisco
- Type
- Full time
- Compensation
- $150k-$200k
- Posted
- 2 days ago
Join a team building useful software. Work closely with engineering and design to turn new ideas into products.
Loading job details
Data Scientist
- Location
- Remote
- Type
- Full time
- Posted
- 4MO AGO
About the role
Open role
Data Scientist
Remote - anywhereFull-timeData & AIAbout the role
The graph is the product. Noded's ingestion chaining™ reads the action layer of 20+ tools - calls, tickets, threads, deals - and turns it into structured, connected customer context that models can actually use. You'll work at the heart of that: how raw activity becomes entities, how entities become insight, and how insight becomes a recommendation a CSM trusts at 8:55 before a 9:00 call.
This is applied science with a short path to production and a very visible bar: our customers see up to 90% lower token costs and measurably better answers because of the work this role owns. You'll make those numbers better - and prove it.
What you'll do
- Advance ingestion chaining: improve how raw, messy, multi-tool activity is parsed, linked, and distilled into the Customer Context Graph.
- Own graph quality: entity resolution across people, accounts, threads, and commitments - the difference between "a note" and "the story."
- Build and evaluate LLM pipelines for insights and recommendations: prompting strategies, retrieval, structured outputs, model selection.
- Make quality measurable: evaluation harnesses, ground-truth sets, and regression gates so the graph only ever gets smarter.
- Push on efficiency: the pre-aggregation and caching strategies behind our token-cost story are a living system - you'll keep extending the lead.
- Partner closely with engineering to take everything above from notebook to production, and with customers to sanity-check that "better" is better for them.
What you'll bring
- A strong applied background - ML, NLP, or data science shipped to production, not just papers and prototypes.
- Hands-on depth with LLMs: prompting, retrieval, evals, and a clear-eyed view of what they're actually good at.
- Solid engineering fundamentals: Python, SQL, and the discipline to build pipelines other people can run.
- Measurement rigor. You're allergic to "it seems better" and know how to design an eval that would change your mind.
- The ability to explain a subtle modeling decision to a non-specialist - clearly, in writing.
Bonus points
- You've worked on knowledge graphs, entity resolution at scale, or recommender systems.
- You've wrangled B2B SaaS data - CRM objects, call transcripts, support tickets - and lived to tell the tale.
- You've published, open-sourced, or written publicly about applied LLM work.
