Agentic AI Architect / Senior AI Architect Company: Innosential AI

Location:
Pittsburgh, Gurgaon, Haryana, India (Hybrid) Employment Type: Full-Time Role Overview Innosential AI is seeking a visionary and technically rigorous Agentic AI Architect / Senior AI Architect to lead the design, architectural governance, and enterprise deployment of autonomous multi-agent systems. In this role, you will architect next-generation agentic workflows, multi-agent orchestration frameworks, and neuro-symbolic guardrails that power secure, production-grade AI solutions for global enterprise clients. You will bridge advanced AI research with robust software engineering, ensuring scalable, low-latency, and zero-trust execution.

Key Responsibilities
* Architect Multi-Agent Systems: Design and implement scalable multi-agent architectures using orchestration frameworks such as LangChain, LangGraph, and Model Context Protocol (MCP) specifications.
* Define Enterprise AI Guardrails: Implement zero-trust AI architectures combining neuro-symbolic guardrails, formal SMT solvers (e.g., Z3), Cedar policy engines, and Trusted Execution Environments (TEEs) for real-time compliance auditing and safety.
* LLM & SLM Integration: Evaluate, fine-tune, and deploy frontier Large Language Models (LLMs) and high-efficiency Small Language Models (SLMs) optimized for specialized domain reasoning and enterprise workflows.
* System Integration & APIs: Design robust integration pipelines connecting agentic workflows with enterprise data stores, REST APIs, microservices, and specialized protocol standards (such as HL7 FHIR for healthcare where applicable).
* Performance & Scalability Optimization: Ensure low-latency inference, fault tolerance, robust state management, and self-healing mechanisms across complex autonomous agent loops.
* Technical Leadership & Mentorship:



Lead cross-functional engineering teams, conduct rigorous code and architecture reviews, and establish engineering best practices for prompt engineering, agent memory management, and deterministic tool-calling. Technical Competencies & Tech Stack
* Agentic Frameworks: LangChain, LangGraph, AutoGen, CrewAI, Model Context Protocol (MCP).
* AI/ML & NLP: PyTorch, Hugging Face Transformers, Vector Databases (Pinecone, Milvus, Qdrant), RAG architectures, prompt engineering, and fine-tuning.
* Security & Governance: Zero-trust security models, formal verification (Z3 SMT solvers), policy engines (Cedar), TEEs, and ZKPs (Zero-Knowledge Proofs).
* Backend & Cloud: Python, TypeScript, FastAPI, Docker, Kubernetes, AWS/Azure/GCP cloud environments.
* Architecture: Microservices, event-driven architectures, API gateways, and distributed state management. Qualifications & Experience
*

Education:
Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Machine Learning, or a related quantitative field.
*

Experience:
8+ years of total software engineering experience and Machine learning experience, with at least 4+ years specifically focused on designing and deploying production-grade LLM applications, generative AI systems, or multi-agent architectures.
* Track Record: Demonstrated portfolio of delivering complex enterprise AI solutions from conception to production.
* Soft

Skills:
Exceptional problem-solving abilities, strong stakeholder communication, and a passion for pushing the boundaries of autonomous agentic systems.

What We Offer
* Chance to spearhead cutting-edge agentic AI innovations at a fast-growing, enterprise-focused AI firm.
* Collaborative culture driven by technical depth, autonomy, and continuous learning.
* Competitive compensation package with performance-based incentives and equity options.

📌 Agentic AI Architect (Bengaluru)
🏢 Innosential
📍 Bengaluru