Senior Engineer, AI

Realty of America; Bengaluru; Hybrid; INR 3,286,000 - 3,756,000; Full time

Job summary

Realty of America (ROA) is hiring a Senior Engineer to lead its AI initiatives. You will decide what to build, implement it, and improve it after deployment. The role covers new products and improving the AI already in use, aiming for faster, cheaper and more accurate outcomes.

You will work to reduce manual effort in real estate workflows, scale AI for agents and support teams, and shape guardrails for safe, reliable AI across systems.

Qualifications

• You have built agents and shipped LLM features to real users.
• You can explain tooling choices and defend your preferences.
• Backend-first mindset with strong Python background.
• Experience measuring results with guardrails and monitoring.
• Open-source contributions are a plus.
• Comfort working with production data and SQL.

Responsibilities

• Explore ideas quickly and bring good ones to production.
• Build agents and LLM-powered features within existing systems.
• Improve live systems on cost, speed and accuracy with evals.
• Ensure output trustworthiness with guardrails and monitoring.
• Share learnings with the team to raise AI-building skills.

Skills

Python backend

Agent frameworks

LLM APIs

Vector search

Model evaluation

Model optimization

LLM monitoring

SQL

AI tooling

Tools

LangGraph

Claude Agent SDK

OpenAI Agents SDK

Job description

Senior Engineer, AI

• $35k – $40k No equity
• Remote India Bengaluru
• 3 years of exp
• Full Time

Job Location

Bengaluru

Remote Work Policy

Onsite or remote

Hires remotely in

India

Visa Sponsorship

Not Available

Preferred Timezones

Pacific Time, Central Time, Eastern Time, Indochina Time

Relocation

Allowed

Skills

• Strong Python backend development
• Building agents with open-source frameworks and harnesses (for example LangGraph, the Claude Agent SDK, the OpenAI Agents SDK or similar)
• Working with LLM APIs: tool use, structured outputs, streaming and prompt design
• Retrieval: embeddings, vector search and RAG pipelines
• Writing evals and measuring model quality, cost and latency
• Making LLM systems cheaper and faster: caching, model routing, batching, choosing smaller models
• Monitoring and tracing LLM apps in production
• SQL and working with real production data
• Daily use of AI coding tools

Nice to have

• Open-source contributions, especially in the AI tooling space
• Model Context Protocol (MCP) servers and tool integrations
• Document understanding: OCR and extraction from forms and PDFs
• Fine-tuning or distilling smaller models
• Enough TypeScript and React to put a prototype in front of users

About the job

About ROA

ROA is a real estate brokerage. We launched about two years ago and have grown quickly since, to around $15M in annual revenue and more than $8B in transactions closed by our agents.

Product and Engineering works a bit like its own company inside ROA, with its own goals and roadmap. Until now, most of our work has been about helping the brokerage run well. More and more, it's about building things that make the industry work better for agents and their clients.

The role

We're hiring a senior engineer to lead our new AI work. You'll help decide what's worth building, build it, and keep making it better once it's live. That includes new products as well as making the AI we already run faster, cheaper and more accurate.

A real estate deal involves a lot of people, paperwork and back-and-forth, and much of it is still done by hand. We think AI can take a lot of that work off people's plates, both for our agents and for the teams that support them. We have the users and the scale to do it well. We need someone who knows how to build it.

The role

We're hiring a senior engineer to lead our new AI work. You'll help decide what's worth building, build it, and keep making it better once it's live. That includes new products as well as making the AI we already run faster, cheaper and more accurate.

A real estate deal involves a lot of people, paperwork and back-and-forth, and much of it is still done by hand. We think AI can take a lot of that work off people's plates, both for our agents and for the teams that support them. We have the users and the scale to do it well. We need someone who knows how to build it.

What you'll work on

• Exploring new ideas quickly, figuring out which ones are worth it, and taking the good ones to production.
• Building agents and LLM-powered features, that fit into the systems we already run.
• Improving what's live: cost, speed and accuracy, with evals to show what actually got better.
• Making AI output trustworthy where it matters, through good guardrails, monitoring and human review.
• Sharing what you learn so the whole team gets better at building with AI.

Who we'd love to meet

• You've built agents yourself, including with open-source harnesses and frameworks, and you've shipped LLM features to real users.
• You have opinions on tooling (models, agent frameworks, eval setups, coding agents) and can explain why you hold them.
• You're a strong backend engineer first. Python is preferred.
• You care about measuring results, not just getting a good demo.
• You like to show your work. When you apply, share something you've built, like a repo, a write-up or a demo.
• Open-source contributions are a big plus.