Sr. Backend Engineer

Dover; New York (NY); On-site; USD 140,000 - 190,000; Full time

Benefits offered by this job

In-person collaboration in NYC

Job summary

Corvus is seeking a senior backend engineer to work directly with the cofounders, owning design through shipping and production monitoring. You will design the data model and architecture for a growing knowledge base about manufacturers, facilities, products, and people, enabling agents to drive research workflows.

You will shape the knowledge graph, decide what to build now vs. later as the customer base expands, and ensure the system scales while maintaining data correctness and consistency.

Qualifications

• 6–10 years of professional software engineering experience with deep relational database expertise.
• Canonical data model design that evolves with new data sources and entity types.
• Experience with entity resolution, deduplication, or reconciliation logic.
• Built resilient systems with queues, retries, idempotency, and observability.
• Familiarity with AI tooling and data structuring for agents, not required to be an AI engineer.
• Pragmatic about scale and able to justify design decisions in plain terms.
• Willing to talk to customers to inform technical decisions.
• Eagerness to collaborate in person with the NYC team, five days a week.

Responsibilities

• Own architecture decisions that shape the system as it grows.
• Design the data model for a growing database of manufacturers, facilities, products, and people.
• Build the knowledge graph and enable agents as primary platform users.
• Scale the system for increasing customer count and data volume.

Skills

Postgres expertise

TypeScript or Python

Data modeling

Entity resolution

Observability

AI tooling familiarity

Customer interaction

Tools

PostgreSQL

TypeScript

Python

Job description

About Corvus

We are tackling a surprisingly unsolved yet critical problem for the $1T chemical manufacturing industry: "what products from my portfolio can I sell into this account"?

A manufacturing salesperson sells hundreds or thousands of SKUs into tens of thousands of manufacturing accounts. To answer the question of what to sell where requires many hours of deep research or in-depth discussion with the customer directly. It also requires either deep chemistry expertise or decades of experience in the space.

Corvus is solving this: by combining agent-powered research workflows with a proprietary, homegrown dataset of US manufacturers, we give our clients understanding about what each one makes, which chemistries go into making it, and how that maps onto a customer's specific portfolio.

That visibility unlocks workflows the industry has never had. Reps find cross-sell opportunities inside accounts they've serviced for years. Sales leaders prioritize markets product by product instead of by gut. Companies see their penetration at the SKU level rather than the account level.

Our confidence in our vision is reflected in our traction: we've 6x'd our revenue, 3x'd our customer count in the last six months, and our initial customers have all renewed and expanded. We work with the largest chemical distributors and manufacturers, and we’re looking to double down for our next phase of growth.

The role

You'll be the senior backend engineer at Corvus, working directly with the two cofounders every day. There's no PM layer: you'll hear the customer's problem firsthand, decide what it means for the system, and own the work from design through shipping and watching how it behaves in production.

Your work will be the foundation of the product. Corvus runs on a growing body of data about manufacturers, facilities, products, and people. Your job is to design the systems and data model around it so that this data stays correct and consistent as it grows, and so that agents and research workflows can build and iterate on top of it. You'll own the architecture decisions that determine what we can build next, and how we scale for our next phase of growth.

Some exciting technical opportunities you'll face:

• Build the knowledge graph for an industry. The clean, current map of data points across hundreds of thousands of manufacturers doesn’t exist. You’ll design the data model that turns all of it into one coherent, queryable picture, providing an unprecedented level of insight and actionability for an entire sector of the economy.

• Make agents the primary users of our platform. Our agents write most of the data and read most of it. There’s no established playbook for this; we will be living on the frontier and defining what "best‑in‑class" looks like.

• Scale the system while the ground is still moving. We’re going from a handful of customers to many. You’ll make the calls on what to build now and what to defer, at the stage where those decisions compound most.

Who we're looking for

• 6–10 years of professional software engineering experience, with deep relational database expertise (preferably Postgres) and strong TypeScript or Python.

• You've designed a canonical data model that lived through real change (new data sources, new entity types, merges and splits) and you can talk concretely about what you'd do differently.

• You've built or owned entity resolution, deduplication, or reconciliation logic, and you understand the tradeoffs between deterministic rules, probabilistic matching, and human review.

• You've built resiliency into systems: queues, retries, idempotency, observability.

• You use AI tooling as a core part of how you build, and you're curious about how data should be structured for agents to work with it. You don't need to be an AI engineer, but the stack shouldn't be foreign to you.

• You're pragmatic about scale. You know what not to build yet, and you can explain your reasoning in plain terms to people who'll push back.

• You want to talk to customers. Our best technical decisions have come from an engineer hearing a customer explain what they'd do next.

• You're excited about working in person with the team in NYC, 5 days a week.

Why Corvus

• Traction. 10x growth over the past year

• Architectural ownership. The data layer is the asset the company is being built on, and you'll be the one who designs it.

• Lean team, high agency. Small team, high bar we set for ourselves, and systems that go from your branch to production without intermediaries.