CLOUDSUFI

What We Are Looking For

CLOUDSUFI is seeking a senior, hands-on AI Platform Architect to design and build production-grade platforms for generative AI, agentic systems, data-intensive applications, and analytical workflows. This is a builder-architect role. The successful candidate will define architecture, make technology decisions, develop reference implementations, review critical code and designs, and guide engineering teams from prototypes to secure, scalable production systems. We are looking for a builder-architect with strong engineering judgement and practical delivery experience. The right candidate can define platform direction, evaluate trade-offs, validate ideas through implementation, and guide systems into production. They should be equally comfortable discussing distributed architecture, reviewing code, diagnosing workflow failures, designing evaluation systems, and mentoring engineering teams.

Key Responsibilities-

AI and Agentic Platform Architecture

• Design platforms for single-agent and multi-agent systems supporting planning, reasoning, tool use, memory, delegation, validation, and human approval.

• Define orchestration patterns for deterministic, dynamic, event-driven, and long-running AI workflows.

• Establish clear boundaries between LLM reasoning, application logic, quantitative computation, rules, and human decision-making.

• Evaluate and adopt agent frameworks, model providers, tools, and orchestration technologies based on reliability, flexibility, performance, and cost. Knowledge and Data Systems

• Architect RAG pipelines, document-processing systems, vector search, hybrid retrieval, knowledge graphs, and semantic data layers.

• Integrate structured and unstructured enterprise data from APIs, databases, files, streams, and external platforms.

• Design reusable workflows for research, data collection, transformation, analysis, modelling, validation, and reporting.

• Establish data lineage, provenance, metadata, access controls, freshness, and quality standards. Evaluation, Observability and Governance

• Build evaluation frameworks for accuracy, relevance, groundedness, task completion, tool use, safety, latency, and cost.

• Enable systematic experimentation across models, prompts, agents, tools, retrieval strategies, and orchestration patterns.

• Implement versioning and lifecycle management for prompts, agents, workflows, datasets, knowledge bases, evaluations, and model configurations.

• Establish tracing, monitoring, auditability, guardrails, approval workflows, and production quality diagnostics.

Cloud and Platform Engineering

• Define cloud-native architectures using microservices, APIs, event-driven systems, queues, schedulers, and distributed processing.

• Lead Kubernetes-based deployment, containerisation, CI/CD, Infrastructure as Code, environment management, and release automation.

• Design for horizontal scalability, fault tolerance, resilience, security, data privacy, and high availability.

• Optimise model usage, infrastructure, storage, retrieval, and compute for performance, latency, and cost.

Technical Leadership

• Translate product and business requirements into clear technical designs and implementation plans.

• Build prototypes and reference implementations for high-risk or foundational platform capabilities.

• Review architecture, code, interfaces, data models, infrastructure, and operational readiness.

• Define engineering standards and reusable patterns across AI, backend, data, and platform teams.

• Mentor senior engineers and support teams in resolving complex technical and production issues.

Required Skills and Experience

• 10+ years of experience in software architecture, platform engineering, distributed systems, data platforms, or AI systems.

• Strong hands-on experience designing and building production-grade AI or data-intensive platforms.

• Deep understanding of LLM applications, tool calling, structured outputs, RAG, embeddings, memory, and agent orchestration.

• Strong experience with cloud platforms, Kubernetes, containers, microservices, APIs, event driven architecture, CI/CD, and Infrastructure as Code.

• Experience with relational, document, graph, vector, and distributed data systems.

• Practical experience implementing AI evaluation, experimentation, tracing, monitoring, guardrails, and lifecycle management.

• Strong understanding of security, identity, access control, secrets management, data protection, and production reliability.

• Ability to move effectively between architecture, code, infrastructure, debugging, and technical delivery.

Good to Have

• Experience building enterprise AI copilots, autonomous workflows, research platforms, or analytical systems.

• Experience with knowledge graphs, hybrid search, model gateways, tool gateways, or agent marketplaces.

• Familiarity with LLMOps, MLOps, model serving, feature stores, model registries, and distributed compute.

• Experience supporting real-time and batch data processing at scale.

• Experience comparing and operating multiple commercial and open-source models.

• Prior experience in consulting, client-facing architecture, or complex enterprise platform delivery. 

About the Role

CLOUDSUFI, a Google Premium Partner specializing in data and AI solutions, is seeking a Staff / Principal Tech Lead to drive the technical execution of our Google Data Commons program. This is a high-visibility, horizontal leadership role embedded within the Google ecosystem — based physically at Google's Bangalore office — working in close daily collaboration with Google's core Data Commons engineering team. The Tech Lead is the technical spine of the engagement. They sit across all four delivery areas — Data Engineering, Frontend, ML/AI, and DevOps/Infrastructure — providing architectural direction, resolving cross track dependencies, and ensuring the quality and coherence of everything we ship. Equally critical is the ability to represent CloudSufi in a credible, articulate, and collaborative manner to Google counterparts at every level.

This is a genuinely hands-on role: the successful candidate must be able to write, debug, and reason about Python and GCP code themselves — not only direct others or lean on AI coding assistants — and must be comfortable operating in an open-source, public-data environment without relying, for example, on Google internal (google3) tooling.

Technology Stack & Domain Knowledge

Core / Must-Have

• Relevant experience on Knowledge Graph, Statistical Data and Analytics (any experience with Google Data Commons is a nice to have, but not required)

• Google Cloud Spanner – schema design, distributed transactions, interleaved tables, and performance tuning at scale.

• Google BigQuery – data modeling, partitioning/clustering strategies, query optimization, and integration with downstream consumers.

• Data pipelines – Apache Beam / Dataflow, or equivalent GCP-native ETL tooling.

• Infrastructure as Code – Terraform on GCP; Cloud Build, Artifact Registry, GKE or Cloud Run.

• Demonstrable, autonomous hands-on proficiency in Python and native GCP tooling — able to code and debug independently in a live technical discussion, with AI-assisted development as a complement to (not a replacement for) that proficiency.

• Direct experience working with open, public, and unstructured datasets (e.g. sourcing, cleaning, and integrating public statistics or open data feeds) using open-source or standard GCP-native tooling.

• Solid grounding in knowledge graph vs data warehouse principles, and how to design for schema and data drift in an open-source knowledge graph context.

• CI/CD experience and working with GitHub

• Full stack experience, specially with data centric apps/systems Strong Advantage

• Python (primary language for Data Commons import tooling and ML pipelines).

• TypeScript / React for the Data Commons web frontend and visualization layers.

• Vertex AI, BigQuery ML, or equivalent ML lifecycle tooling.

• Knowledge graph principles, RDF/SPARQL, or statistical data modeling.

• DataCommons Python / REST APIs and the DCID import automation tools.

• Experience with Google's internal engineering culture, tools (e.g. Buganizer, Critique, Cider), or prior delivery inside a Google product or partnership engagement.

Experience & Qualifications Required

• 10+ years of software engineering experience, with at least 3 years in a formal or informal tech lead capacity overseeing multiple workstreams.

• Demonstrable experience delivering production-grade systems on Google Cloud Platform.

• Prior experience working with or for Google — as a Googler, through a Google partnership program, or as a contractor embedded in a Google team — is strongly preferred.

• Exceptional verbal and written English communication skills; able to engage confidently with senior Google engineers and program managers.

• Proven ability to operate across ambiguous, fast-moving programs with multiple parallel tracks.

• Based in Bangalore, India, and able to work on-site at Google's Bangalore office on a regular basis.

• Able to read and interpret an RFP / SOW and connect its terms to a workable technical delivery plan.