Solution Architect

Hyderabad, Telangana

Job Summary

What you’ll do Discover and shape the solution Run discovery workshops with client stakeholders, assess the current process, and frame the business problem as an agentic solution.Prioritise and shape use cases jointly with the client, then validate direction with a working prototype in days — followed by the production hardening that enterprise delivery demands. Build and integrate Design and ship single- and multi-agent systems in Python using ADKs, connecting into client ecosystems through proven integration patterns, standards-based APIs and secure authentication.Engineer the agent core: prompt engineering, context engineering, prompt caching, RAG / context-graph retrieval, memory, tool / function calling, MCP integration and multi-agent orchestration.Build reusable agent components, skills, tool libraries and templates, with guardrails so agent behaviour stays predictable, safe and repeatable. Prove and govern Run eval-driven development: build evaluation harnesses and test suites that measure correctness, safety and regression before anything ships.Own AgentOps and DevSecOps — CI/CD for agents, versioning, observability, telemetry, IAM, secrets management, privacy controls and shift-left security — with Responsible AI governance from day one.Take agents through architecture review, security sign-off and release governance on the way to production, then hand over with clear documentation and knowledge transfer into managed service.Design human oversight into every agent: approvals, exception handling and escalation paths appropriate to the process being automated. Own the outcome Set a baseline with the client, agree benefit targets, and track cycle time, manual effort, accuracy, exception rate, cost-to-serve, adoption and compliance as the scorecard.Run a continuous feedback loop: feed production telemetry, evals and client feedback back into prompts, context and agent design. Operate within HCLTech and the client Collaborate with account, consulting, architecture, data, security and delivery teams — the FDE role sits across both organisations, not apart from them.Keep internal alignment as tight as client alignment: follow HCLTech’s governance charter,and socialise and review the proposition with the wider SME cohort before committing the client to it.Look for the scaling opportunity in every engagement — where a pattern, accelerator or approach could extend across other domains within the Delivery Unit — and bring proactive proposals forward. Lead and mentor Set technical direction for the pod: run code and design reviews, mentor engineers, plan the work, manage delivery risk and raise the engineering bar.Stay ahead of the curve: adopt evolving agent frameworks and patterns quickly, and bring field learnings back to the practice.

Key Responsibilities

Technology mandate Language Python — primary and mandatory; other languages and frameworks follow the engagement. For example in the ING context Java will be necessary. Frameworks Agent Development Kits (ADKs) — e.g. Google ADK, LangGraph, CrewAI, OpenAI Agents SDK, AWS Bedrock AgentCore, Microsoft Agent Framework / Semantic Kernel. Framework choice follows the engagement; the discipline is the same. Models Multi-LLM via the kit (e.g. Claude on Bedrock, Gemini, Azure OpenAI), selected per use case for quality, latency and cost. Interfaces Tools and Model Context Protocol (MCP) for integration; standards-based APIs and secure auth for client systems.

Skill Requirements

What you’ll bring (must-have) Advanced Python engineering — idiomatic, typed, tested and packaged code — on a solid software engineering foundation (design, version control, architecture). Other languages follow the engagement; Python is the core.Hands-on agent building with at least one ADK (Google ADK, LangGraph, CrewAI, OpenAI Agents SDK, AWS Bedrock AgentCore or Microsoft Agent Framework / Semantic Kernel).Solid command of agent engineering: prompt engineering, context engineering, prompt caching, RAG / context graphs, tool / function calling, MCP and multi-agent orchestration.Eval-driven development: designing evaluation harnesses and measuring agent quality, safety and reliability.Standards-based integration and DevSecOps: APIs, secure authentication and authorisation, IAM, CI/CD, observability, resilience and AgentOps.Ability to conceptualise a business problem as an agent quickly, and operate effectively in ambiguous, client-embedded settings.Client-facing consulting skills: discovery facilitation, process assessment, use-case selection, stakeholder management and executive communication.Comfortable working across account, consulting, architecture, data, security and delivery teams to get a solution to production.Responsible AI as a working discipline — applied throughout agent design, evaluation and production governance.Experience mentoring or leading small engineering teams.

Other Requirements

A day in the life A typical day blends client consulting with hands-on engineering. It often starts alongside business and technology stakeholders — unpicking a process bottleneck, agreeing what a good outcome looks like, and shaping it into an agentic solution. From there the work turns practical: building or refining an agent, wiring it into client systems, reviewing evaluation results and production telemetry with the engineering pod, and working through security controls and Responsible AI checks as delivery issues surface.Throughout, there’s a leadership thread — setting direction for the pod and raising the engineering bar — alongside the constant balancing act of speed against enterprise-grade quality: showing the client visible progress, measuring business impact, folding in user feedback, and turning what works into reusable HCLTech accelerators.