Solution Architect

Job Summary We are looking for a hands-on Solution Architect to own the technical design of client-facing and internal platforms across .NET, modern JavaScript frontends, and cloud. You will translate business requirements into architectures that are secure, scalable, and economically sensible then stay close enough to the code to make sure they are actually built that way. This is not a whiteboard-only role. You will write code, review it, unblock teams, and defend your decisions in front of both engineers and clients. You will also help shape how we adopt AI/ML and LLM-based capabilities across our delivery portfolio. Key Responsibilities Architecture & Design Own end-to-end solution architecture: application, integration, data, and deployment layers. Translate business and functional requirements into architecture blueprints, HLDs/LLDs, sequence diagrams, and interface contracts. Make and document technology decisions with explicit trade-offs (ADRs) including the options rejected and why. Define service boundaries, API contracts, data models, and integration patterns across systems. Design for non-functional requirements from day one: performance, scalability, availability, security, observability, and cost. Drive modernisation of legacy applications (e.g., .NET Framework / .NET Core current LTS) with realistic, staged migration paths. Set and enforce engineering standards: coding guidelines, branching strategy, code review expectations, test strategy, and definition of done. Hands-On Engineering Build proofs of concept, reference implementations, and reusable accelerators for delivery teams. Review critical code and pull requests; intervene directly on complex or high-risk components. Diagnose production issues — performance bottlenecks, memory pressure, N+1 queries, cache design, concurrency problems. Champion CI/CD, infrastructure-as-code, and automated testing as default practice, not afterthoughts. Cloud & Platform Design and govern cloud architecture on Azure or AWS, including compute, storage, networking, identity, and secrets management. Right-size and cost-model solutions; present TCO and cost-optimisation options to stakeholders. Define containerisation and orchestration strategy where appropriate (Docker, Kubernetes/AKS/EKS). Establish observability standards — logging, metrics, tracing, alerting — and make sure teams actually adopt them. AI/ML & LLM Solutions Assess where AI/ML genuinely adds value and, equally important, where it does not. Architect LLM-backed features: RAG pipelines, chunking and embedding strategy, vector store selection, retrieval quality, prompt and context design, and orchestration. Understand and articulate the trade-offs between prompt engineering, RAG, and fine-tuning for a given problem. Design for the practical constraints of LLM systems: latency, token cost, rate limits, caching, fallback behaviour, and graceful degradation. Define evaluation and guardrail approaches — accuracy/groundedness measurement, hallucination mitigation, PII handling, and human-in-the-loop checkpoints. Collaborate with data engineering on the pipelines and data quality these systems depend on. Client & Pre-Sales Participate in client solutioning discussions, technical workshops, and discovery sessions. Contribute to proposals, effort estimation, solution approach documents, and RFP responses. Present and defend architecture to client stakeholders — technical and non-technical alike. Support technical due diligence and client architecture review boards. Leadership & Mentoring Mentor senior and mid-level engineers; raise the technical bar of the teams around you. Support technical hiring — interview design, panel participation, and candidate evaluation. Contribute to internal knowledge sharing, tech talks, and reusable IP. Required Skills Must Have — Depth Expected 8–12 years in software engineering, with a demonstrable track record of architecting and shipping production systems. Strong .NET / C# — ASP.NET Core, Web API, Entity Framework Core, async programming, dependency injection, and current LTS versions. One cloud platform in depth — Azure or AWS. Azure preferred (App Service, Functions, AKS, Azure SQL / PostgreSQL, Service Bus, Key Vault, Entra ID, Application Insights). AWS equivalents equally acceptable. At least one modern frontend framework in depth — Angular or React — with solid working knowledge of the other, plus TypeScript. Node.js for services, tooling, or BFF layers. Databases: strong SQL (SQL Server / PostgreSQL), data modelling, indexing, and query tuning. Working knowledge of NoSQL and when to reach for it. Architectural patterns: microservices, event-driven architecture, messaging/queues, API gateway, CQRS, caching strategies — including honest judgement on when a monolith is the better answer. Security: OAuth2/OIDC, JWT, RBAC, secrets management, OWASP Top 10, secure SDLC. DevOps: Azure DevOps or GitHub Actions, CI/CD pipeline design, IaaC (Bicep/ARM/Terraform), Docker. Must Have — Working Understanding Practical understanding of AI/ML concepts and the ML lifecycle. Hands-on exposure to LLM-based application development — RAG architectures, embeddings, vector databases (pgvector, Azure AI Search, Pinecone, or similar), and orchestration frameworks (LangChain, Semantic Kernel, or equivalent). Familiarity with major model providers and the trade-offs between hosted APIs and self-hosted/open models. Python for AI/ML and data work. Non-Technical Clear written communication — you will be judged on your documents as much as your designs. Ability to explain a technical trade-off to a CFO and to a senior engineer, on the same day, without changing your answer. Comfortable being challenged and equally comfortable changing your mind when the argument is better. Ownership: you follow a decision through to production, not to the end of the design phase. Qualification Bachelor's degree in Computer Science, Engineering, or a related field. Master's is a plus. Equivalent demonstrable experience will be considered in place of formal qualifications. Preferred (Added Advantage) Cloud architect certification — Azure Solutions Architect Expert (AZ-305) or AWS Certified Solutions Architect – Professional. Experience in an IT services or consulting environment with multiple concurrent clients. Exposure to data engineering and analytics — Azure Synapse, Databricks, PySpark, dbt, Power BI. Experience with Kubernetes in production. Experience delivering an AI/LLM feature to production, including its evaluation and cost model. Open-source contributions, technical writing, or conference speaking. What Success Looks Like in This Role First 90 days Mapped the current architecture landscape and identified the top technical risks with an evidence-backed view. Delivered at least one architecture decision record and one working reference implementation. First 6–12 months Engineering standards defined, documented, and visibly adopted by delivery teams. Measurable improvement in delivery predictability, defect escape rate, or cloud cost. At least one AI/LLM capability designed, evaluated, and delivered to production or a credible pilot. Two or more engineers visibly levelled up under your mentorship. What We Offer Genuine architectural ownership across a portfolio of client and internal platforms. Direct access to leadership and a short path from decision to implementation. Freedom to shape our AI and cloud engineering practice rather than inherit someone else's. Sponsorship for certifications and continuous learning. A collaborative, engineering-led environment. How to Apply Share your CV at mrunal.kanade@scriptshub.net with the subject line "Application – Solution Architect – Pune" . Include a brief note on one architecture decision you made that turned out to be wrong, and what you did about it.