Senior Agentic Engineer, AI Platform

About the role

This position sits at the intersection of distributed systems and generative AI, owning the core platform that powers agentic workflows at scale. You will be responsible for designing and implementing the low-latency infrastructure that enables autonomous agents to execute complex, multi-step tasks reliably. The role demands deep collaboration with product and security teams to translate data protection policies into automated, runtime enforcement within the agentic loop. You will own the architecture that ensures sensitive data is processed in compliance with customer mandates while maintaining high throughput. A significant portion of this role involves building self-healing mechanisms that allow the platform to automatically remedize detected violations without human intervention. You will also architect the observability and debugging tools that provide transparency into agent decision-making and execution traces. Ultimately, this role empowers security teams to automate risk management, allowing end-users to focus on innovation rather than compliance overhead.

Key facts

What you'll do

Architect and maintain the real-time data ingestion pipelines that feed agentic workflows, ensuring high volume event streaming via Kafka is processed with sub-second latency.

Design and operate the semantic and classic search infrastructure that powers agentic retrieval, optimizing index performance and accuracy at massive scale.

Build and scale the auto-scaling compute layer that executes highly concurrent agentic workflows while optimizing for cost, latency, and LLM accuracy.

Implement automatic remediation workflows that resolve security violations in production before they escalate into customer incidents.

Develop observability and monitoring solutions that provide granular insights into agent behavior, system performance, and customer usage patterns.

Lead the architecture of internal mission-critical databases and services that store and manage sensitive customer configurations and metadata.

Write and maintain comprehensive documentation for internal and public services, ensuring clear handoffs and long-term maintainability.

Partner with security engineers to embed data loss prevention policies directly into the agentic execution fabric.

Refactor legacy components to support modern deployment patterns, improving reliability and developer velocity.

Define and track platform-level SLIs and SLOs to ensure the AI platform meets stringent reliability targets.

Collaborate with product teams to prototype new agentic capabilities and validate technical feasibility for new customer requests.

Optimize data processing workflows using Kafka, Flink, and Snowflake to handle evolving compliance and auditing requirements.

Evaluate and integrate emerging tooling for agentic evals, ensuring the platform continuously improves its autonomous decision quality.

Champion best practices for secure coding, ensuring every service handles authentication, authorization, and secrets management correctly.

Requirements

Expertise in one or more systems/high-level programming languages (e.g. Python, Go, Java, C++) and the eagerness to learn more.

Experience running scalable (thousands of RPS) and reliable (three 9's) systems in production environments.

Solid experience with building Search and Agentic AI systems at massive scale: ingestion, text summarization, semantic and classic search, and agent orchestration. Experience with evals is a plus.

Experience with developing complex software systems scaling to substantial data volumes or millions of users with production quality deployment, monitoring, and reliability.

Experience with large-scale distributed storage and database systems (SQL or NoSQL, e.g. Postgres, Cassandra) to manage persistent state and high throughput demands.

Ability to decompose complex business problems and lead a team in solving them through technical design and code reviews.

Data Processing - experience with building and maintaining large scale and/or real-time complex data processing pipelines using Kafka, Flink, Snowflake, Hadoop or Zookeeper.

8+ years of experience with proven technical leadership and prowess in scalable, distributed architectures, demonstrating a history of shipping critical infrastructure.

TOOLS WE USE

• Go
• Node.js http://node.js/
• React
• Python
• Cassandra
• Redis
• Terraform
• Docker
• Kubernetes
• AWS
• Kafka
• Envoy

Environment

Nightfall AI takes pride in being an equal-opportunity employer. We value a diverse and global talent pool and the collaboration that results from having a diverse and inclusive team. All applicants will be considered for employment without attention to race, color, religion, sex, sexual orientation, gender identity, national origin, veteran or disability status. Our hiring decisions are based exclusively on merit, qualifications, and business needs.

Compensation

Employee compensation will be determined based on interview performance, level of experience, specialization of skills, and market rate. During the offer discussion, your recruiter will review the finalized base salary, bonus (for applicable roles), benefits & perks, and stock options as they'll be reflected in the offer letter.

Practical notes

The role is full-time based in Bengaluru.