Founding AI Engineer - Manhattan Labs

Praxis, Inc.

San Francisco (CA)

On-site

USD 140,000 - 220,000

Full time

14 days+

Get more replies from employers

Send a job-specific resume in minutes.

Benefits offered by this job

Competitive salary and equity

Health, dental, vision coverage

Commuter stipend

Regular team events and off-sites

Job summary

Manhattan Labs, in San Francisco, is seeking a founding engineer to architect and build an AI-native platform for regulatory compliance. You’ll shape the system, own critical technical decisions, and partner with the founding team to push our enterprise-ready vision from MVP to production.

You’ll design agentic reasoning, large-scale document ingestion, and user-facing tools while ensuring regulatory scrutiny and reliability for financial institutions.

Qualifications

1+ years of experience on technical projects with startup experience strongly preferred
Strong TypeScript and Python skills across modern frontend and backend stacks
Hands-on experience shipping LLM integrations or RAG systems into production user-facing products
Experience with document processing, entity extraction, and structured data extraction from messy real-world sources
Solid grasp of API design, relational databases, and async job processing systems
Proven track record taking a product from "works locally" to reliable production at growing scale
Debugging instincts across frontend, backend, infra, and data boundaries - and an ownership mindset to match

Responsibilities

Design, build, and operate RAG and agentic systems over complex financial data, with clear evaluation metrics for accuracy, latency, and cost
Develop scalable backend services for document processing (PDF/OCR/CSV), async job queues, case lifecycle orchestration, and risk scoring with confidence-driven UX and human-in-the-loop controls
Own the productionization of AI systems, including evaluation pipelines, dataset/version management, and safe rollout strategies (A/B testing, guardrails, fallback behavior)
Harden the platform for enterprise: auth controls, role-based access, audit trails, backups, and monitoring with measurable SLOs
Establish CI/CD, deployment pipelines, runbooks, and incident response practices from the ground up
Propose and engage in cutting-edge AI research related to our mission, particularly around agentic reasoning over regulatory and financial data
Work directly with the founding team to define architecture, technical roadmap, and product direction

Skills

TypeScript

Python

LLM integrations

Document processing

API design

Async job processing

Ownership mindset

Startup experience

Frontend/backend stacks

Tools

LangChain

LlamaIndex

Job description

About the team

We’re building the AI-native compliance platform for community banks and financial institutions.

Regulatory pressure is intensifying. Financial crime is becoming more sophisticated. Compliance teams are overwhelmed by manual workflows and fragmented systems. Community banks, despite managing billions in assets, lack the tooling to keep up.

We’re building the intelligence layer that changes that: a system that can answer questions like
“What is our exposure to this entity across all customers?”
- with a defensible, auditable answer.

Our team includes engineers and researchers from Stanford, Carnegie Mellon, and IIT Madras, and we’re backed by Tier 1 investors and experienced operators who understand the massive opportunity in modernizing financial crime compliance with AI.

Learn more at https://manhattanlabs.co/

Description of the role

This is a true founding engineer role. You won’t just ship features - you’ll define the system.

You’ll architect and build an AI-native platform that operates as the investigative brain for compliance teams. This includes agentic reasoning systems, large-scale document ingestion, and user-facing tools that must stand up to regulatory scrutiny.

You will take the product from early-stage MVP to a secure, reliable, enterprise-grade system used by financial institutions.

By reimagining KYB and AML as autonomous intelligence, you'll help build the platform that becomes the default compliance infrastructure for thousands of community banks, that each manage assets worth billions of dollars. You'll work directly with the founding team, own technical decisions, and lay the foundation for a potentially massive and fast-growing SaaS company.

Responsibilities and potential impact

Design, build, and operate RAG and agentic systems over complex financial data (corporate registries, sanctions lists, beneficial ownership, adverse media), with clear evaluation metrics for accuracy, latency, and cost

Develop scalable backend services for document processing (PDF/OCR/CSV), async job queues, case lifecycle orchestration, and risk scoring with confidence-driven UX and human-in-the-loop controls

Own the productionization of AI systems, including evaluation pipelines, dataset/version management, and safe rollout strategies (A/B testing, guardrails, fallback behavior)