Full-time. Founding team. New York City, in-person required. Reports to the Co-founder & CTO and works directly with both founders and domain experts.
About the company
We bring that information together so teams can understand the evidence, apply their own judgment and preserve why they made a decision.
Our ambition is a world model for corporate governance: a system that connects institutional knowledge, policies, decisions and outcomes. Getting there starts with reliable data and software customers trust in their daily work.
Responsibilities
Build and own the data systems that make the company useful and trustworthy.
This is a backend- and data-intensive applied AI role. You should be as comfortable debugging a production pipeline and evolving a database schema as evaluating a document-extraction model. You will build on an existing codebase, working with messy filings and customer documents and making the results dependable enough for customers to use. The work includes asynchronous jobs, versioned evidence and safe reprocessing, not just prompts and model experiments.
Your first mandate is one prioritized pipeline and its downstream use. You will establish what good looks like with domain experts, improve the system, and own it in production. As one of our first engineering hires, you will also help shape how we build, test and operate software.
What you'll own
Build and improve ingestion and extraction pipelines for public filings, external feeds and customer documents. Choose structured sources, deterministic parsing or models based on the task. Make jobs safe to retry, prevent duplicate or stale-worker writes, recover from partial failures, and make missing or stale coverage visible.
Design data models for the entities and relationships your pipeline needs. Preserve source evidence, effective dates and version history so past outputs remain reproducible as data, policies and models change. Distinguish verified facts, model outputs, human decisions and missing information.
Build evaluations with domain reviewers using independently human-reviewed reference cases. Measure completeness, citation support, consequential errors and cases requiring human judgment, not just aggregate agreement. Treat model-generated references as proposals, not ground truth, and turn production failures into regression tests.
Provide reliable services for research, policy application and recommendations. Make reprocessing safe: retain customer corrections, identify affected outputs and flag results that need renewed review.
Protect customer information throughout ingestion, storage, evaluation and reprocessing. Respect tenant boundaries and permitted data uses. Do not assume one customer's documents or decisions can be reused for another.
Operate what you build. Own monitoring, diagnosis and tested recovery for your services. Manage provider usage, caching and processing costs without hiding gaps or sacrificing quality. Reduce manual intervention and document enough that another engineer can release and support them.
You will work closely with our product engineering counterpart, who owns the customer workflow and how people inspect, correct and use the outputs. You own the underlying pipeline and its data-quality contract. Agree on interfaces and failure behavior together, and follow issues through to resolution rather than stopping at a handoff.
The founders set priorities and resolve tradeoffs. Founders and domain reviewers provide policy interpretation and reviewed reference answers; you are not expected to invent governance rules yourself. We will scope your initial work so you can make progress even if the other role has not yet been filled.
Problems you might tackle
Two disclosures disagree about a director's committee membership. Determine which facts apply at which dates, preserve both sources and make the discrepancy reviewable.
A model update improves average extraction accuracy but introduces a consequential error. Catch it with an evaluation and make a defensible release decision.
A new filing arrives after an analyst has approved an output. Update the relevant facts without losing their corrections or silently carrying approval over to changed results.
Your first 90 days
By day 30: Trace one agreed pipeline from source to customer output, ship an improvement, and establish quality, coverage and operating baselines with domain reviewers.
By day 60: Own that pipeline in production, with reviewed test cases, monitoring, safe reprocessing and a tested recovery procedure.
By day 90: Demonstrate an agreed improvement in reliability, coverage or review effort without sacrificing the quality baseline. Routine releases and diagnosis no longer need a founder to direct each step, and another engineer can operate the service using your documentation.
We will choose the initial pipeline and success measures together based on customer needs and the existing system.
What you bring
Strong backend engineering, SQL and Postgres fundamentals, with experience operating production services or data pipelines.
Practical experience with asynchronous or distributed processing: idempotency, retries, leases and fencing, concurrency, and recovery when only part of a job succeeds.
Experience turning messy documents or other imperfect source data into structured information people depend on. Here that means public company filings, third-party data feeds and customer spreadsheets.
Practical experience shipping model-based systems and improving them through evaluations against human-reviewed references, user feedback and failure analysis, with the observability to explain what a model did on a given input.
Sound judgment about schema evolution, entity resolution, historical records and data provenance.
The ability to become productive in an existing codebase, debug across services and storage systems, and evolve interfaces or schemas without breaking their consumers. Our services are TypeScript and Go over Postgres and BigQuery.
The independence to narrow an ambiguous problem, ask for the context you need and carry the work through production use.
Experience with financial or legal documents and human-review tools is useful. Governance expertise and model-training research are not prerequisites. We care more about systems you have made reliable than a particular language, model or framework.
Our stack
TypeScript services on NestJS and Go services over gRPC, with Postgres as the system of record and BigQuery for analytical and document data. We use Anthropic and OpenAI models with LangSmith tracing, run on Google Cloud with Kubernetes and Terraform, and observe with Datadog and Sentry. CI enforces coverage gates. Python is useful for eval and analysis work.
Conditions
We work in person in New York and stay close to customers. We prototype quickly, use AI development tools where they help, and remain responsible for what we ship.
We narrow scope before compromising correctness, permissions or customer trust. We test representative cases and failure paths, observe what happens after release, and flag risks early. Founding engineers have room to make decisions and are expected to make their reasoning understandable to the team.
Apply
Send a short note and examples of systems you have built. Describe a difficult data or model failure, how you traced it to its source, what you changed and how you verified the fix. Confidential work can be discussed without sharing proprietary code or customer data.