Senior AI Software Development Engineer

F5 F5 · Enterprise · Hyderabad, India +1

Senior AI Software Development Engineer at F5, focused on leading the implementation, rollout, and optimization of enterprise-scale agentic AI transformation. This role involves developing scalable agentic systems, orchestration frameworks, and integrating enterprise AI platforms like Gemini and Vertex AI. Responsibilities include building agent lifecycle management, enabling high-code agent development, ensuring observability, safety, and performance, and supporting engineering teams in adopting AI development platforms.

What you'd actually do

  1. Build and optimize enterprise-grade agent orchestration frameworks supporting tool use, memory, RAG, agentic workflows and automation.
  2. Integrate Gemini models via Vertex AI, ensuring secure, scalable API consumption and robust model routing.
  3. Develop reference implementations and reusable libraries for high-code agents in Java, Python, Go, or TypeScript.
  4. Build logging, tracing, telemetry, and evaluation pipelines for agent performance and reliability.
  5. Optimize inference latency, parallelization, and cost management strategies across agent workflows.

Skills

Required

  • 7+ years of experience in software engineering
  • strong experience in distributed systems and backend architecture
  • Deep hands-on coding expertise in Python
  • hands-on production experience with LLM-based systems
  • prompt engineering
  • tool calling
  • RAG
  • embeddings
  • agent frameworks
  • Experience with Vertex AI, Gemini APIs, OpenAI APIs, or similar enterprise AI platforms
  • Strong understanding of API design, microservices, and cloud-native architectures (Docker, Kubernetes)
  • Experience building or integrating orchestration frameworks (e.g., LangChain, LlamaIndex, custom orchestration layers)
  • Familiarity with vector databases, embedding pipelines, and retrieval strategies
  • Strong understanding of authentication, authorization, and enterprise security patterns
  • Proven ability to build robust, reusable code and clean APIs

Nice to have

  • Experience building multi-agent systems or autonomous workflow engines
  • Experience with model evaluation pipelines and AI quality metrics
  • Familiarity with structured output enforcement (JSON schemas, function calling)
  • Experience working with enterprise data systems such as Snowflake, Salesforce, ServiceNow, SharePoint
  • Knowledge of cost modeling and inference optimization techniques
  • Experience contributing to internal developer platforms or SDK ecosystems
  • Go
  • Java
  • TypeScript

What the JD emphasized

  • high-code agent development and orchestration frameworks
  • production-grade AI agents
  • production code
  • enterprise-grade agent orchestration frameworks
  • tool use
  • memory
  • RAG
  • agentic workflows
  • automation
  • multi-agent collaboration
  • event-driven execution
  • workflow chaining
  • agent lifecycle management
  • state persistence
  • context engineering
  • Gemini models
  • Vertex AI
  • secure, scalable API consumption
  • robust model routing
  • internal SDKs
  • reusable components
  • prompt engineering
  • token efficiency
  • grounding strategies
  • structured output patterns
  • high-code agents
  • Java
  • Python
  • Go
  • TypeScript
  • Salesforce
  • Snowflake
  • SharePoint
  • ServiceNow
  • internal APIs
  • MCP (Model Context Protocol) servers
  • secure API integration gateways
  • logging
  • tracing
  • telemetry
  • evaluation pipelines
  • agent performance
  • reliability
  • safety guardrails
  • input/output validation
  • hallucination mitigation
  • prompt injection defenses
  • policy enforcement
  • Security
  • secure data handling
  • RBAC enforcement
  • compliance alignment
  • Gemini Code Assist
  • CLI workflows
  • internal AI development platforms
  • technical documentation
  • internal libraries
  • code samples
  • code reviews
  • AI-enabled applications
  • inference latency
  • parallelization
  • cost management strategies
  • agent workflows
  • caching strategies
  • streaming responses
  • batching techniques
  • throughput
  • performance testing
  • benchmarking
  • agents
  • models
  • different workloads
  • distributed systems
  • backend architecture
  • Python
  • Go
  • Java
  • TypeScript
  • LLM-based systems
  • prompt engineering
  • tool calling
  • RAG
  • embeddings
  • agent frameworks
  • Vertex AI
  • Gemini APIs
  • OpenAI APIs
  • enterprise AI platforms
  • API design
  • microservices
  • cloud-native architectures
  • Docker
  • Kubernetes
  • orchestration frameworks
  • LangChain
  • LlamaIndex
  • custom orchestration layers
  • vector databases
  • embedding pipelines
  • retrieval strategies
  • authentication
  • authorization
  • enterprise security patterns
  • robust, reusable code
  • clean APIs
  • multi-agent systems
  • autonomous workflow engines
  • model evaluation pipelines
  • AI quality metrics
  • structured output enforcement
  • JSON schemas
  • function calling
  • enterprise data systems
  • Snowflake
  • Salesforce
  • ServiceNow
  • SharePoint
  • cost modeling
  • inference optimization techniques
  • internal developer platforms
  • SDK ecosystems
  • secure, scalable orchestration layer
  • enterprise agents
  • production-grade high-code agents
  • clean, reusable components
  • robust observability
  • evaluation
  • safety controls
  • AI-driven workflows
  • AI agent rollout
  • integrations
  • SDKs
  • ready-to-use reference implementations
  • agentic AI
  • engineering
  • sales
  • support
  • finance
  • product teams
  • core technical solutions
  • pro-code AI development
  • enterprise-grade reliability
  • governance
  • security

Other signals

  • enterprise-scale Agentic AI transformation
  • lead the implementation, rollout, and optimization of high-code agent development and orchestration frameworks
  • develop scalable agentic systems
  • build and optimize enterprise-grade agent orchestration frameworks supporting tool use, memory, RAG, agentic workflows and automation
  • Implement logging, tracing, telemetry, and evaluation pipelines for agent performance and reliability
  • Apply safety guardrails including input/output validation, hallucination mitigation, prompt injection defenses, and policy enforcement