AI Engineer

About Astra:

Astra is a cyber security SaaS company that makes otherwise chaotic pentests a breeze with its one of a kind Pentest Platform. Astra's continuous vulnerability scanner emulates hacker behavior to scan applications for 9300+ security tests. CTOs & CISOs love Astra because it helps them fix vulnerabilities in record time and move from DevOps to DevSecOps with Astra's CI/CD integrations.

Astra is loved by 650+ companies across the globe. In 2023 Astra uncovered 2 million+ vulnerabilities for its customers, saving customers $69M+ in potential losses due to security vulnerabilities.

We've been awarded by the President of France Mr. Franois Hollande at the La French Tech program and Prime Minister of India Shri Narendra Modi at the Global Conference on Cyber Security. Loom, MamaEarth, Muthoot Finance, Canara Robeco, ScripBox etc. are a few of Astras customers.

Role Overview

At Astra Security, we are building the future of AI-driven cybersecurity, including Autonomous Pentesting- an experience that helps teams automatically discover, test and secure Web & Mobile applications, Cloud and APIs at scale.

As an AI Engineer, you will design and build AI-powered Agentic systems that can reason, retrieve knowledge and execute complex workflows autonomously. You will work closely with the engineering and product teams to build AI-native applications that integrate with Astras platform and security scanners.

This role focuses on building AI-based Agentic applications, implementing RAG architectures, and orchestrating AI workflows across distributed systems. You will work with frameworks like LangGraph or Mastra to build production-grade AI systems powered by frontier LLM models and self-hosted open-source models.

Here's what's in it for you

AI System Development

• Build AI-powered agentic applications that can reason, plan and execute multi-step workflows autonomously.
• Design and develop AI solutions using frameworks such as LangGraph, Mastra or similar frameworks.
• Integrate LLM models from OpenAI, Anthropic or open-source self-hosted LLMs into production systems.
• Work closely with platform and backend teams to integrate AI capabilities across Astras cybersecurity products.

RAG & Knowledge Systems

• Design and implement Retrieval Augmented Generation (RAG) architectures for contextual reasoning.
• Build and maintain AI knowledge bases that allow agents to retrieve relevant context and make better decisions.
• Develop pipelines for embedding generation, indexing and semantic retrieval.
• Continuously optimize prompts, retrieval strategies and system performance.

Agentic Workflow Orchestration

• Build multi-agent workflows capable of orchestrating tasks across distributed systems.
• Implement agent execution pipelines that integrate with Astras scanners, APIs and platform services.
• Ensure AI workflows are reliable, scalable and production-ready.

Collaboration & Innovation

• Work closely with engineering leaders and product teams to ship AI-powered features.
• Experiment with new AI frameworks, models and architectures to improve product capabilities.
• Contribute to building Astras AI-first security platform

AI Systems Engineering

• Build production-grade AI services with proper observability, logging and monitoring.
• Design APIs and backend services that expose AI capabilities to Astras platform.
• Implement caching, rate limiting and cost optimization strategies for LLM workloads.
• Work with vector databases (ChromaDB, PGVector or others) to build scalable retrieval systems.
• Build structured tool interfaces that allow agents to interact with APIs, scanners and internal services.

AI Evaluation & Reliability

• Design evaluation frameworks to measure accuracy, hallucination rates and reasoning quality of AI agents.
• Build automated evaluation pipelines for RAG and agent workflows.
• Implement guardrails, structured outputs and validation layers to ensure reliable AI behavior.
• Improve system performance through prompt engineering, tool design and retrieval optimization.

Tools & Technologies

• Python or TypeScript (preferred)
• LangGraph, Mastra, CrewAI or similar agent frameworks
• Vector databases (FAISS, PGVector, Pinecone, Weaviate)
• LLM APIs (OpenAI, Anthropic, Mistral, Llama)
• Docker & containerized deployments
• Git and CI/CD pipelines
• Basic cloud experience (AWS / GCP)

Nice to Have

• Experience building RAG systems in production
• Experience with graph-based knowledge systems (Neo4j, GraphRAG)
• Experience building multi-agent systems
• Familiarity with cybersecurity concepts or vulnerability management
• Experience deploying AI services on cloud infrastructure

Location

This is a hybrid role in Bengaluru and requires a regular in-office presence to collaborate closely with Engineering and Product teams. Being in-office helps foster deep collaboration, faster decision-making and the shared energy of building together - values that define how we operate at Astra.

Required Qualifications & Skills

• 2+ years of experience in software engineering, AI engineering or applied ML roles.
• Candidates should have built at least one end-to-end AI application (RAG system, AI agent, or LLM-powered application) that integrates with external APIs or services.
• Hands-on experience building AI-based agentic applications.
• Experience with frameworks like LangGraph, Mastra or similar frameworks.
• Strong understanding of RAG (Retrieval Augmented Generation) systems and building knowledge bases.
• Experience orchestrating agentic workflows in distributed systems environments.
• Experience building AI solutions using LLMs from OpenAI, Anthropic or open-source/self-hosted LLM models.
• Strong programming skills in Python or TypeScript.
• Experience integrating LLMs into production applications.
• Strong problem-solving skills and ability to build scalable AI systems.

Why Astra?

• We're reimagining cybersecurity with products like the API Security Platform and Autonomous Pentesting, helping teams discover, test, and secure Web, Mobile, Cloud, and APIs at scale.
• Youll be joining a team of builders and problem-solvers who thrive on first-principles thinking, move fast without breaking trust and take pride in shipping products that matter.
• We work closely with founders, embrace ownership at every level and value curiosity as much as expertise.
• If youre excited about solving hard problems, scaling products globally and shaping the future of cybersecurity - Astra is the place for you.
• Annual trips to beaches or mountains (last one was at Wayanad).
• Health insurance & other benefits