Agent Engineer
Bengaluru • FullTime
Posted 25d ago
About the job
The distance between an AI system that works and one that millions of people rely on every day is where the real engineering lives. Crossing it is what Agent Engineers at Sarvam do. The agents you build go into live operation at India's largest banks, insurers, NBFCs and government departments, and at some of the country's fastest-growing companies, handling millions of real interactions, measured against outcomes the customer wants to drive. We treat agents as code, not prompts someone tweaks in a console, but engineered artifacts: versioned, reviewed, and held to a regression suite, where you can reason about what a change will do and roll it back if you are wrong. Sarvam is India's leading full-stack AI company, building sovereign AI infrastructure and applications purpose-built for India.
Responsibilities
- Build production agents on the Sarvam stack, including prompts, system design, workflows, tool integrations, and evals.
- Map the space an agent needs to cover, identifying critical scenarios and edge cases for designed coverage.
- Build proofs of concept to demonstrate functionality to customers before large-scale commitment.
- Integrate agents into diverse customer systems such as core banking, CRM, ticketing, and telephony.
- Own end-to-end quality by defining metrics, building evaluation pipelines, and ensuring performance.
- Engineer agents with software development discipline, including versioning, code reviews, and regression testing.
- Tune agents for production constraints like accuracy, cost, reliability, and performance under load.
- Design for scale and consistency, ensuring systems maintain behavior and are well-instrumented.
- Adapt proven agent patterns to new customer environments efficiently and with high quality.
- At senior levels, own technical build playbooks and quality bars, and mentor partner engineers.
Requirements
- Shipped to production and owned the system afterwards, including maintenance and issue resolution.
- Experience working seriously with LLMs and agents, including context engineering, tool use, structured outputs, memory, and large-scale failure modes.
- Experience with agentic frameworks and both open and closed-source models.
- Applied engineering discipline to AI, focusing on versions, tests, regressions, and reproducibility.
- Proficiency in defining relevant metrics, building evaluation pipelines, and distinguishing signal from noise.
- Strong Python skills and broad experience across the stack, including backend services, data pipelines, APIs, and cloud infrastructure.
- Ability to quickly learn and adapt to unfamiliar technology stacks.
- Designed systems for long-term behavior, predictability under load, visibility through instrumentation, and understandability by others.
- Ability to collaborate effectively with customer engineering teams, understanding constraints and reaching workable solutions.
- Minimum of 3 years of experience, with a focus on building and maintaining systems used by real users.