Software Engineer (AI products & data)

Indigo is building the AI Operating System for real estate. Within months of our first pilot, we already process more > 300,000 documents a month and > 1% of all US home transactions. And we've yet to start marketing and press.

• We're a second-time founding team whose last company reached $100M annualized revenue in the same industry.
• We're building cutting edge technology that no one else in real estate has seen before, to power a scalable platform with multiple products on top of it.
• We're building infrastructure so AI can handle the tedious parts of real estate, so that people can enjoy the fun parts.
  

We're hiring an exceptional backend‑leaning engineer who can design and ship elegant, reliable services.

• Backend work will be your primary lane (services, data, queues, durable execution), but you’re comfortable jumping into data pipelines, infra, and the occasional frontend task when the team needs it. Of course, you'll have claude-code/codex/AI tools to help.
• You’ll push the boundaries of LLM‑powered backends (prompt/tooling infra, evals, latency/cost controls) while holding a high bar for software craftsmanship.
• We use TypeScript on both frontend and backend. We don’t expect you to have experience in everything we use, but we do expect strong TypeScript fundamentals and experience building robust, scalable systems.

Fewer years of experience? Just apply. We care about how impactful you are, not how many years you've worked.

• Ideal location is in NYC. Our engineering and product team is in NYC, where we go to an office 3-5d a week in Soho.
• We're open to hiring in the East Bay/Bekerley, CA where one of our founding engineers is based.
  

• Design and ship backend services in TypeScript/Node NestJS that power agent, team, and brokerage workflows
• Build reliable data pipelines (batch and streaming) to ingest, normalize, and validate large volumes of documents and events (OLTP→OLAP handoffs, schema evolution, idempotency)
• Architect event-driven systems with queues to orchestrate long-running workflows
• Own LLM operations: prompt/tool versioning, function/tool use, caching, cost/latency budgets, evaluation harnesses, and guardrails
• Develop offline/online evals for model quality (golden sets, canary traffic, regression alerts) and wire them into CI/CD and runtime telemetry
• Expose clear, stable APIs (GraphQL/REST) and evolve them safely with contracts and compatibility tests
• Instrument services for observability (tracing/metrics/logging), reliability (SLOs, retries/backoff, dead-letter handling), and performance
• Collaborate across functions (eng, product, design, ops, GTM) to translate customer feedback & fuzzy requirements into real products
  

• T-shaped generalist with depth in backend systems; breadth across data engineering, infra/DevOps, and enough frontend to collaborate and debug
• TypeScript/Node experience building production backends and APIs
• Experience building agents with LLMs in a production env is a huge plus
• Quick & avid learner: you learn new domains, technologies quickly, powered by your intellectual curiosity
• Founder-like mindset: you want to win, you take full ownership from problem thru impact, and you build with tenacity and resourcefulness
• Strong data engineering fundamentals: data modeling, ETL/ELT, stream processing, and storage trade-offs (Postgres/OLTP, columnar/OLAP)
• Hands-on experience with queues and workflow orchestration (SQS/SNS; familiarity with durable execution patterns)
• Solid testing discipline (unit/integration/property tests), CI/CD, and observability
  

• TypeScript across the stack; NestJS services; GraphQL (Apollo Client/Federation) for APIs
• We use tons of AWS services, Postgres, Clickhouse, Kafka, BullMQ, Redis, etc.
• Observability: Sentry, CloudWatch, tracing/metrics
• LLM providers & tools: Closed & open sourced models, LangSmith
  

• Small, highly talented teams with high ownership and low coordination overhead will run circles around larger, mediocre teams because of the power law of software. Even more true with AI coding tools.
• Quality compounds—well-designed software ships faster over time. Leverage AI as much as possible, but apply taste so you get speed not slop.
• AI and tools are improving at breakneck pace. Curiousity helps us keep up with the cutting edge and knowing what works vs what doesn't yet.
• Broad skillsets amplify impact: the same person doing full-stack, data, AI/LLM, dev-ops dramatically increases impact and speed. Also true beyond engingeering skills–product & business sense give you judgement to build the right things
• We’re building something huge and which takes a lot of hard work and time. We support each other to sustain the pace (many of us have families—flexibility matters)