Lead Backend Engineer

Stage 2 Capital; Delhi; On-site; INR 4,000,000 - 6,000,000; Full time

Job summary

Stage 2 Capital is seeking a lead backend engineer to own and harden the core systems of our product. You will work in a Python backend with clear service boundaries, delivering production changes with high ownership, and elevating reliability, performance, and design.

You will shape ranking and retrieval quality, multi-tenant data models, and complex external API integrations, including LLM-driven flows. This role emphasizes deep ownership, robust testing, and collaboration across teams.

Qualifications

• 7+ years building production backend systems.
• Expert-level Python and comfortable in large codebases.
• Strong experience with relational data modeling (PostgreSQL).
• Experience with search or information retrieval (Elasticsearch/OpenSearch).
• Experience designing reliable integrations with third-party APIs (auth, rate limits, webhooks).
• Familiarity with async job systems (Celery, Sidekiq, RQ) and production debugging.
• High bar for code quality: tests, interfaces, and careful rollouts.
• Ability to own problems end-to-end with minimal hand-holding.

Responsibilities

• Own and harden backend systems at the centre of the product.
• Ship production changes with high ownership and reliability.
• Improve ranking/retrieval quality and multi-tenant data models.
• Build and orchestrate LLM-driven QA and tool-use flows.
• Develop integrations & auth flows (OAuth/SSO/SAML) across platforms.
• Maintain production systems with observability, testing, and robust deployments.

Skills

Python

PostgreSQL

Elasticsearch/OpenSearch

Async programming

Kubernetes

GCP

React/TypeScript

OAuth/SAML authentication

Unit/integration testing

Code quality & design

Tools

Airflow

Celery

dbt

Redis

Job description

The Role

We’re hiring a lead backend engineer to own and harden the systems at the center of our product. You’ll work in a Python backend with clear service boundaries, ship production changes with high ownership, and raise the bar on reliability, performance, and design.

This is not a CRUD API role. You’ll reason about ranking and retrieval quality, multi-tenant data models, external API complexity, and LLM systems that have to work under real support-desk constraints.

What You'll Work On

• Search & retrieval: Modern search and retrieval systems (e.g., Elasticsearch, PgVector), airflow, filters, embeddings, and result quality


• Answer / QA services: orchestration around LLMs, prompt engineering, and agentic tool use


• Integrations & auth: connectors and OAuth/SSO/SAML flows across various support platforms


• Async & data pipelines: Async job systems, data pipelines, and ingestion workflows for heterogeneous source data


• APIs & data model: REST services, ORM/data modeling, and database migrations, multi-tenant correctness


• Production systems: Redis, container orchestration (e.g., Kubernetes), observability, feature flags, and robust testing frameworks

What We're Looking For

• 7+ years building production backend systems (or equivalent depth)


• Expert-level Python and comfort in large existing codebases


• Strong experience with relational data modeling (Postgres), migrations, and query performance


• Experience with search or information retrieval (Elasticsearch/OpenSearch, vector search, or similar): or clear evidence you can ramp fast


• Experience designing reliable integrations with third-party APIs (auth, rate limits, webhooks, eventual consistency)


• Familiarity with async job systems (Celery, Sidekiq, RQ, or similar) and production debugging


• High bar for code quality: tests, clear interfaces, careful rollouts, written design when the problem warrants it


• Ability to own problems end-to-end with minimal hand-holding

Must Have

• LLM/RAG systems in production (grounding, evaluation, tool calling)


• Flask or comparable Python web frameworks (Django/FastAPI fine)


• Enterprise identity (SAML, OAuth2, SSO)


• Airflow, Singer/Meltano taps, or dbt


• Kubernetes / GCP


• Frontend TypeScript, React for full-stack debugging


• Familiarity with ticket systems (e.g., Salesforce, Zendesk, ServiceNow)

How We Work

• Collaborative, high-ownership engineering culture


• Ship in small, reviewable PRs; pair when the problem is hard


• Depth matters: we value engineers who can go deep on search quality, integration edge cases, and production incidents

What Success Looks Like In The First 6 Months

• Independently ship backend features across search, answers, or integrations with clean designs and solid tests


• Become a trusted owner for at least one critical backend surface (e.g. retrieval, QA orchestration, or a major connector family)


• Improve reliability or performance of a hot path you’ve touched


• Raise team quality through thoughtful reviews and clear technical writing