Software Engineer III / Staff Engineer – Agentic AI (Bengaluru)

Software Engineer III / Staff Engineer – Agentic AI (Bengaluru)

19 Nov
|
Quabrick labs search
|
Bengaluru

19 Nov

Quabrick labs search

Bengaluru

Software Engineer III / Staff Engineer – Agentic AI

Location: Bengaluru

Experience: 6+ Years (SDE III) | 9+ Years (Staff Engineer)

Build AI Agents That Work in the Real World!

We are looking for exceptional backend engineers who have hands-on experience building, deploying and scaling LLM-powered AI agents in production.

This is not a conventional backend engineering role or an opportunity to simply integrate LLM APIs. You will solve challenging engineering problems around agent architecture, tool calling, multi-step reasoning, evaluation, reliability and production scalability.

You will own critical parts of the agent lifecycle — from architecture and development to evaluation, deployment, monitoring and continuous improvement.

What You Will Build & Own

- Design and build production-grade LLM agents involving tool calling, multi-step execution and orchestration.
- Own agent development end-to-end, including prompts, tools, context management, memory, guardrails and deployment.
- Build robust agent evaluation frameworks using golden datasets, automated evaluations, LLM-as-judge and human feedback.
- Establish measurable quality benchmarks using task success rate, precision, recall, F1, tool-call accuracy and regression rates.
- Design scalable backend services and distributed systems to support reliable AI agent execution.
- Drive production observability, tracing, failure analysis, retries, timeouts, latency and cost optimisation.
- Continuously improve agent performance by diagnosing real-world failures and implementing measurable improvements.

Non-Negotiable Requirements

- Strong Backend Engineering: 6+ years (SDE III) / 9+ years (Staff) building and owning production backend systems.
- Production Agentic AI Experience: At least 1 year of hands-on experience building and shipping LLM agents used by real customers.
- End-to-End Ownership: Personally built and owned at least one LLM agent across design, tools, evaluation, deployment, monitoring and post-launch improvements.




- Evaluation-Driven Development: Must demonstrate concrete evaluation evidence — a named metric, dataset size, evaluation framework or release gate.
- Robust System Design: Deep understanding of distributed systems, backend architecture, APIs, scalability and engineering fundamentals.
- Production Debugging: Ability to explain actual agent failures encountered in production and how they were diagnosed and fixed.

Additional Expectations for Staff Engineers

- Drive agent architecture and engineering standards across multiple agents or teams.
- Establish reusable evaluation frameworks, golden datasets and release standards.
- Implement evaluation-gated releases, CI checks, shadow/canary deployments and rollback mechanisms.
- Make sound architectural decisions around model selection, cost, latency, context, memory and orchestration.
- Remain hands-on with coding, code reviews and technical mentorship.

Good to Have

- Experience with LangGraph, LangChain, MCP, CrewAI or similar agent frameworks.
- Exposure to LangSmith, Braintrust, DeepEval, Arize Phoenix or custom evaluation frameworks.
- Experience with multi-agent systems, model/provider migration and LLM observability.
- Strong Python, Java or other backend programming experience.
- Experience building AI systems in high-scale or regulated environments.

Why Join?

- Solve challenging problems at the frontier of Agentic AI.
- Build and scale intelligent systems used in real-world production environments.
- Influence architecture, engineering practices and the future of AI development.
- Work with a high-ownership engineering team building next-generation AI systems.

If you have built AI agents that have survived real production workloads — and can demonstrate how you measured, debugged and improved them — we would love to hear from you! Pay: ₹6,500,000.00 - ₹9,500,000.00 per year

Experience

- Agentic AI : 1 year (Required)
- Platform Building : 3 years (Required)

Location:
- Bangalore City, Bengaluru, Karnataka (Required)

Work Location: In person

📌 Software Engineer III / Staff Engineer – Agentic AI (Bengaluru)
🏢 Quabrick labs search
📍 Bengaluru

Reply to this offer

Impress this employer describing Your skills and abilities, fill out the form below and leave Your personal touch in the presentation letter.

Subscribe to this job alert:

Get the latest job offers by email for: software engineer iii / staff engineer – agentic ai (bengaluru) / bengaluru

Subscribe to this job alert:

Get the latest job offers by email for: software engineer iii / staff engineer – agentic ai (bengaluru) / bengaluru