Software Engineer 3, AI Framework Integrations

AI frameworks like LangChain, LlamaIndex, and n8n are quickly becoming the default way developers build with AI — and MongoDB wants to be the data platform that shows up everywhere they build. As part of the AI Builders Experience (ABX) org, we're standing up a brand new engineering team in Gurugram to make that happen, and we're looking for Software Engineers to be among its founding members.

The team owns the connective layer between MongoDB and third-party AI frameworks, platforms, and tools — the integrations that let developers use MongoDB effectively with the AI stack they've already chosen. That means shipping into some of the fastest-moving open-source ecosystems in tech, working mostly in Python and TypeScript, and partnering with the Database Experience (DBX) org when an integration needs new driver, platform, or product capability underneath it.

This is a strong role for an engineer who wants to grow into the AI space with real ownership. You'll own well-scoped projects from design through release, work with fast-moving and sometimes unproven technologies, and get a clear view of how the ecosystem is evolving.

Priorities shift as frameworks change and customer demand moves, so we're looking for engineers who are adaptable, ask good questions, and take pride in shipping.

Much of your work will happen in public: sending pull requests to upstream repositories, working through external maintainer review cycles, and participating in developer communities as MongoDB.

You'll work closest with the engineers sitting next to you, pairing on hard problems, reviewing each other's code, and dividing up a broad portfolio, with senior engineers and your manager close at hand as you take on larger pieces of it.

MongoDB engineering teams pride themselves on building high-quality software and living our cultural values every day. We value intellectual curiosity, intellectual honesty, customer empathy, and building together in an environment that prioritizes collaboration over competition.

We are looking to speak to candidates who are based in Gurugram for our hybrid working model.

Position expectations

Build, test, and ship MongoDB integrations across a broad AI framework ecosystem that includes technologies like LangChain, LangGraph, LlamaIndex, n8n, and Mastra

Own well-scoped projects end to end on weekly-to-quarterly timelines, taking responsibility for the scope, schedule, and quality of your own work and communicating progress and trade-offs clearly

Solve day-to-day technical problems with growing independence, using the team's established patterns and past practice to guide your decisions, and seeking guidance on non-standard problems

Contribute to exploratory work on new frameworks: help evaluate an integration's feasibility, build a first working version of the piece you own, and share what you learn with the team

Contribute high-quality code upstream to repositories MongoDB doesn't own, following the team's contribution practices and maintaining a high standard of professionalism and courtesy through external maintainer review cycles

Work within and improve the team's engineering practices: CI/CD, automated testing across framework and server versions, version compatibility checks, and release steps. Suggest improvements where you see repeated friction

Help keep the integrations you work on healthy: monitor for upstream breaking changes, handle upgrades, triage incoming issues, and follow through after release

Coordinate with adjacent teams, including DBX, when your work depends on driver, platform, or product capability, and escalate cross-team dependencies and risks early rather than working around them

Collaborate with product management, product design, and other partners to understand what users need from an integration and how your work connects to the team's objectives

Write clear documentation, give useful code reviews, share what you know with teammates, and follow through on your commitments

Develop specialized knowledge of the Python or TypeScript AI builder ecosystems — frameworks, agent and retrieval patterns, vector search — and bring interesting developments back to the team

Qualifications

3+ years of experience building backend systems, libraries, or other developer-facing software, with solid proficiency in Python and/or TypeScript

Experience designing, writing, testing, and debugging software in a shared, long-lived code base, with an eye for the tests and documentation that make it maintainable

Track record of delivering scoped projects reliably and on time, with growing independence and good judgment about when to ask for help

Demonstrated ability to work on several projects at once and keep quality and commitments intact as priorities shift

Clear interest in modern AI builder workflows and the developer tooling landscape: AI frameworks, agentic patterns, RAG and retrieval, embedding, or vector search

Solid verbal and written technical communication, including the ability to write things down clearly for teammates and partners you won't always overlap with in real time

Enjoys collaboration and being part of a close-knit team; is approachable, curious, and intellectually honest

Eager to learn, with a strong technical foundation and an interest in deepening it

Nice to Have

Open source contributions, especially any work that landed in a repository you don't own

Experience building AI or framework integrations, MCP servers, agent skills, or plugins for agentic applications, professionally or as side projects

Familiarity with the agentic AI tooling ecosystem (AI IDEs, CLIs, and assistants) and how developers integrate with it

Experience with CI/CD and automated testing for libraries that support multiple versions or platforms

Experience working with databases; familiarity with MongoDB query patterns

Exposure to vector search, embedding pipelines, or index management

Working knowledge of Java, Go, C#, or Rust

Experience with containerization tools (e.g. Docker, Kubernetes) and with AI frameworks and libraries

Familiarity with Jira and related tools

Success measures

In the first month, you will understand the team's scope, how our integrations are structured, where the relevant repositories and pipelines live, and how we build, test, and release. You'll have landed your first code contribution and know who to ask when you're blocked

In three months, you will be delivering scoped work on one or more integrations with growing independence, will have landed an upstream contribution, and will be comfortable running the team's testing and release process for your own changes

In six months, you will own a well-defined area or project end to end, deliver against quarterly objectives with limited oversight, and have made at least one concrete improvement to the quality or efficiency of how the team works

In twelve months, you will be a dependable owner of a piece of the portfolio, will have delivered multiple meaningful features or integrations, will be handling upstream and customer issues in your area with confidence, and will be helping teammates get up to speed on what you know best