Senior AI Engineer, Agentic Data Enrichment

Baselayer; San Francisco (CA); Hybrid; USD 230,000 - 340,000; Full time

Benefits offered by this job

Flexible PTO

In-person in SF 4 days / week

Health, dental, vision coverage

401(k) with match

Gym stipend $250/mo

Job summary

Baselayer is hiring a Senior AI Engineer to own a slice of the enrichment surface end-to-end. You will build production-grade LLM-driven agents that classify businesses, verify web presence, and link officers to entities, while optimizing cost and latency.

You’ll work in a hybrid SF setting, owning model selection, prompt design, and evaluation across multiple LLM providers. Join a foundational team building a real-time entity graph used by financial institutions.

Qualifications

• Shipped LLM-driven agents to production.
• Strong async Python including structured data libs, web frameworks, and relational DBs.
• Experience across multiple frontier LLM providers and at least one agent framework.
• Built or maintained eval methodology: golden datasets, scoring, labeling guidelines.

Responsibilities

• Own industry/category classification from signals.
• Develop risk/legitimacy scoring from web signals.
• Link individuals to businesses via public web evidence.
• Select models, design agents, prompts, tools, and evaluation methods.
• Manage cost and latency for enrichment surface.

Skills

LLM agents

Async Python

LLM providers

Eval methodology

JSON schema vs functions

Tools

LangSmith

Braintrust/Evals

Job description

Every business in America needs a bank account to exist. The system that decides whether they're real, who's behind them, and whether they're a risk, runs on infrastructure from the 1980s. We're rebuilding that layer from scratch.

Baselayer is the identity layer for institutions across the United States — the most complete business graph in America and every human tied to it. We fuse public records, IRS data, sanctions lists, web signals, and fraud telemetry from 2,200+ financial institutions into a single graph that resolves any business and the humans behind it in milliseconds. The legacy credit bureaus took 50 years to build something that gets 60% match rates. We've built something that gets 98% in under two years.

Today we're trusted by over 20% of financial institutions in America — including FIS, Rho, Socure and leading loan infrastructure providers. But the graph is becoming infrastructure for anyone who needs to know if a business is real and worth trusting: gig platforms, marketplaces, AI companies, and commerce infrastructure at scale.

Trust is the substrate of every financial transaction. We're rebuilding it.

ABOUT THE TEAM

We're solving real-time entity resolution at a scale no one else has cracked — fusing dozens of data sources into a single business identity graph and resolving any entity in milliseconds. It's a graph AI problem, a retrieval problem, and a fraud-modeling problem stacked on top of each other. The technical depth is real.

You'd be joining a small team where the data moat is defensible, the research problems are open, and the infrastructure you build becomes load‑bearing for businesses. Ownership is real. Velocity is real. There's no layer of process between an idea and shipping it.

We're at an inflection point — the graph is built, the match rates speak for themselves, and the hardest problems are still ahead: graph embeddings, fraud propagation models across the business network, real‑time traversal at sub‑100ms latency, and expanding the identity layer beyond finance into every platform that needs to trust a business.

If you want to work on something foundational — the kind of infrastructure that gets built once and everything else runs on top of — this is it.

ABOUT THE ROLE

Baselayer answers questions the loan application didn't ask. For every business that crosses our queues, we need to know things that aren't on the form: what the business actually does, where it actually lives on the web, whether the people it names match the public record, and whether anything across the open web contradicts the story we were told. We answer those questions with LLM‑driven agents that crawl, click, search, and extract structured evidence from across the web - and we treat this as a production data pipeline, not a research demo. We're hiring a Senior AI Engineer to own a slice of this enrichment surface end‑to‑end.

WHAT YOU'LL DO

• Own industry/category classification of businesses from heterogeneous signals (name, website, directory presence, reviews).
• Build and maintain discovery and verification systems for a business's real web presence - filtering aggregators, parked domains, brand collisions, and impersonators.
• Link individuals to businesses via public web evidence (e.g. confirming a named officer or employee genuinely works there).
• Develop risk/legitimacy scoring derived from web‑presence signals, fed back into downstream underwriting.
• Own model selection, agent design, prompt and tool engineering, eval methodology, and cost control across your enrichment surface.

MINIMUM REQUIREMENTS

• Shipped LLM‑driven agents to production - not notebooks, not demos. Real users, real cost, real failure modes, real on‑call.
• Strong async Python including structured-data libraries, modern web frameworks, and relational databases.
• Experience across multiple frontier LLM providers and at least one agent framework, with deep knowledge of failure modes.
• Built or maintained eval methodology: curated golden datasets, scoring functions, labelling guidelines, regression diagnostics.
• Holds informed opinions on structured‑output reliability - when to use JSON‑schema mode vs. function calling vs. extractor‑on‑top‑of‑text.

WHAT SETS YOU APART

• Web scraping at scale: anti‑bot evasion, residential proxies, request fingerprinting, authenticated flows, CDN defeats.
• Eval‑framework experience (e.g., LangSmith, Braintrust, Evals, or custom).
• Entity resolution / record linkage / fuzzy matching at scale.
• Built a tool registry or toolset abstraction over multiple LLM providers.
• Cost/latency optimization: response caching, semantic caching, model routing (cheap‑first then elevate), thinking‑budget tuning, prompt‑cache hit‑rate work.

WORK LOCATION

• Based in SF; hybrid - 4 days per week in office.

COMPENSATION

• Salary Range: $230,000 – $340,000 + Equity

BENEFITS

• Time off when you need it: Flexible PTO so you can recharge without red tape.
• In‑person energy: We're based in SF and meet in the office 4 days a week.
• Competitive compensation: We pay well and back it with equity. We want you to think and act like an owner.
• Career rocket fuel: You'll help build the foundation of a high‑growth startup, working side by side with experienced founders and team members who've done it before.
• Benefits on us: We cover 100% of your health, dental, and vision premiums. No surprise deductions from your paycheck.
• 401(k) with company match: We match your contributions so your future self benefits too
• HSA contributions included: We contribute to your HSA on applicable plans, so your coverage works as hard as you do
• Stay healthy, stay sharp: A $250 monthly gym stipend to help you bring your best self to work, and everywhere else
• A seat at the table: We believe in transparency, radical candor, and giving every team member a voice