AI Architecture Professional

Agilysis; Hyderabad, Pune District, Bengaluru; On-site; INR 3,000,000 - 6,000,000; Full time

Job summary

Agilysis is seeking a hands-on AI Architect to design and guide secure, scalable AI solutions across enterprise workflows. You will connect business needs with practical architecture, advise teams on model/platform choices, and support solutions from discovery through production adoption while collaborating with Product Owners.

You will lead architecture, agent orchestration, RAG pipelines, and governance across systems, ensuring cost-effective, compliant, and observable AI deliverables in a

Qualifications

• Owns production AI/ML solutions and enterprise integrations.
• Understands LLM behavior, prompting, context limits and evaluation trade-offs.
• Experience implementing tool-enabled agents and RAG pipelines.
• Ability to measure quality, latency, cost, and ensure secure access controls.
• Able to translate business value into actionable AI roadmaps and collaborate with cross-functional teams.

Responsibilities

• Define target architectures and phased delivery plans across ML/AI workflows.
• Design agents with tool calls and workflow orchestration.
• Architect document ingestion, embeddings, vector search, and grounded responses with citations.
• Provide technical leadership, diagrams, and reviews for security and engineering stakeholders.
• Evaluate models and platforms for privacy, latency, and total cost; optimize tokens and routing.
• Integrate AI services with existing applications, databases, and APIs.
• Define governance, data handling controls, and responsible AI practices.
• Establish production readiness with deployment pipelines and observability.

Skills

Enterprise architecture

LLMs and AI

Agents and RAG

Evaluation and security

Business & communications

Tools

Vector databases

Observability tools

Model gateways

Job description

Job Summary

We are looking for a hands-on AI Architect to design and guide the delivery of secure, scalable AI solutions across enterprise workflows. Who can connect defined business needs with practical architecture, advise teams to choose appropriate models and platforms, and support taking solutions from discovery through production adoption while working closely with Product Owners.

Responsibilities

• Architecture and strategy. Assess business workflows, data readiness and existing systems. Define target architectures and phased delivery plans, choosing among rules, traditional ML, RAG and agent-based approaches according to the use case.
• AI agents and orchestration. Design agents with tool calling and workflow orchestration. Experience with agent orchestration frameworks, vector databases, enterprise search, model gateways and LLM evaluation or observability tools.
• RAG and enterprise knowledge. Architect document ingestion, parsing, chunking, embeddings, vector and hybrid search, reranking and grounded responses with source citations. Preserve source permissions and keep knowledge current.
• Technical leadership. Produce architecture diagrams, integration contracts and decision records. Review designs and prototypes, mentor agent developers, and explain technical trade-offs to business, security and engineering stakeholders.
• Model selection and economics. Evaluate models and platforms against task quality, privacy, latency, deployment needs and total cost. Optimize token usage, context size, caching and model routing; support in measure cost per successful task and workflow ROI.
• Legacy integration. Connect AI services to existing applications, databases, APIs, batch processes and messaging systems.
• Governance and responsible use. Define practical AI do's and don'ts, approved usage patterns, data-handling controls and review gates. Address sensitive data, prompt injection, unreliable outputs, auditability and human oversight.
• Production readiness. Define supporting cloud or hybrid infrastructure, identity and access, networking, secrets, deployment pipelines and observability. Establish evaluation gates, service targets, monitoring and incident response.

Required experience and skills

• Enterprise architecture delivery. Demonstrated ownership of production AI or ML solutions and enterprise integrations, including architecture decisions, implementation guidance and operational outcomes.
• LLMs and traditional AI. Strong understanding of large language model behavior, prompting, context limits, embeddings, model evaluation and fine-tuning trade-offs, alongside established ML methods and their lifecycle.
• Agents and RAG. Hands-on experience implementing tool-enabled agents and RAG pipelines. Ability to explain retrieval quality, grounding, tool execution, memory boundaries, failure modes and appropriate human intervention.
• Evaluation and security. Ability to define representative test datasets and measure answer quality, retrieval relevance, agent task completion, latency and cost. Experience implementing access controls, privacy protections and traceable operations.
• Business and communication. Ability to estimate business value, challenge unsuitable AI use cases, guide cross-functional teams from proof of concept to production and ability to work in fast-paced implementation.

Preferred experience

• Knowledge of HR processes and enterprise HR platforms such as Workday, including workflow integration, employee-data access and approval boundaries.
• Experience in complex or regulated enterprises with legacy modernization, sensitive information and formal architecture governance.