ML Solutions Architect

As an ML Solutions Architect, you’ll be the technical bridge between clients and delivery teams. You’ll lead pre-sales technical discussions, design ML architectures that solve business problems, and ensure solutions are feasible, scalable, and aligned with client needs. This is a highly client-facing role requiring both deep technical expertise and strong communication skills.

In the era of Generative AI and autonomous systems, you’ll also be responsible for architecting agentic solutions that leverage LLMs, tool ecosystems, and AI-assisted workflows to deliver transformative value to clients.

Core Responsibilities Pre-Sales and Solution Design (45%)

• Lead technical discovery sessions with prospective clients;
• Understand client business problems and translate them into ML solutions;
• Design end-to-end ML architectures and technical proposals;
• Create compelling technical presentations and demonstrations;
• Estimate project scope, timelines, cost, and resource requirements;

• Support General Managers in winning new business. Client-Facing Technical Leadership (25%)

• Serve as the primary technical point of contact for clients;
• Manage technical stakeholder expectations;
• Present technical solutions to both technical and non-technical audiences;
• Navigate complex organizational dynamics and conflicting priorities;
• Ensure client satisfaction throughout the project lifecycle;

• Build long-term trusted advisor relationships. Agentic Solutions Architecture (15%)

• Architect agentic AI solutions that leverage autonomous decision-making and tool orchestration;
• Design MCP (Model Context Protocol) integration strategies for client environments;
• Evaluate and recommend appropriate agent frameworks (LangGraph, Claude Agent SDK, etc.) for client use cases;
• Create POC demonstrations showcasing agentic capabilities using AI-assisted development tools
• Advise clients on build vs. buy decisions for agentic components;
• Develop reference architectures for common agentic patterns (RAG agents, multi-agent systems, tool-using agents);

• Assess AgentOps requirements including monitoring, evaluation, and cost optimization. Internal Collaboration and Handoff (15%)

• Collaborate with delivery teams to ensure smooth handoff;
• Provide technical guidance during project execution;
• Contribute to the development of reusable solution patterns and agentic accelerators;
• Share learnings and best practices with ML practice;
• Mentor engineers on client communication and solution design;

• Contribute to Provectus AI toolkit documentation and solution templates., Success Metrics (First 90 Days) Days 1-30:

• Shadow 3-5 pre-sales engagements;
• Build relationships with General Managers and sales team;
• Complete onboarding to Provectus solution catalog and AI toolkit;
• Contribute to at least 1 technical proposal;

• Demonstrate proficiency with Claude Code and AI-assisted development for POC creation. Days 31-60:

• Lead 2-3 technical discovery sessions independently;
• Create compelling technical demonstrations including agentic AI capabilities;
• Successfully hand off 1-2 projects to delivery teams;
• Build rapport with key clients;

• Develop at least one reusable agentic solution pattern or reference architecture. Days 61-90:

• Win at least 1 new client engagement through technical leadership;
• Establish yourself as trusted technical voice for agentic AI solutions;
• Contribute to at least 1 reusable solution asset or AI toolkit component;
• Receive positive feedback from clients and internal stakeholders;
• Successfully architect and propose at least one agentic solution to a client.

What We Offer

• High-visibility role working with diverse clients;
• Opportunity to shape solution offerings and practice direction;
• Work with cutting-edge ML, LLM, and agentic AI technologies;
• Global exposure across LATAM, Europe, and North America;
• Career path toward Practice Leadership or Principal Architect;
• Learning budget and conference attendance;
• Remote-first with regular client travel opportunities;
• Access to latest AI tools and subscriptions for professional development.

Application Process

1. Initial Screening: Resume + preliminary discussion;
2. Technical Assessment: System design + ML/agentic architecture discussion (2 hours);
3. Case Study: Design solution for realistic client scenario including agentic components (presentation);
4. Behavioral Interview: Client communication and stakeholder management;
5. Final Round: Meet with Practice Leader and General Managers;
6. Offer.

• Solution Design: Ability to architect end-to-end ML systems for diverse business problems;
• ML Lifecycle: Deep understanding of the full ML lifecycle from data to deployment;
• System Design: Experience designing scalable, production-grade ML architectures;
• Trade-off Analysis: Ability to evaluate technical approaches (cost, performance, complexity);
• Feasibility Assessment: Quickly assess if ML is an appropriate solution for a problem., * Agentic Architecture: Deep understanding of agent design patterns, state management, and orchestration frameworks;
• Claude Ecosystem: Hands-on experience with Claude Code, Claude Agent SDK, and Anthropic’s tool ecosystem;
• MCP Proficiency: Understanding of Model Context Protocol architecture for designing client integrations;
• Agent Frameworks: Practical knowledge of LangGraph, LangChain agents, and multi-agent orchestration patterns;
• AI-Assisted Workflows: Demonstrated experience with AI coding assistants (Cursor, GitHub Copilot, Claude Code) for rapid prototyping;
• Tool Ecosystem Design: Ability to architect function calling and tool use strategies for complex client requirements;
• AgentOps Understanding: Knowledge of agent monitoring, evaluation frameworks, and cost optimization strategies;

• POC Development: Ability to rapidly build compelling agentic demonstrations using AI-assisted development. ML Breadth

• Multiple ML Domains: Experience across various ML applications (RAG, Computer Vision, Time Series, Recommendation, etc.);
• LLM Solutions: Strong experience in architecting LLM-based applications including agentic systems;
• Classical ML: Foundation in traditional ML algorithms and when to use them;
• Deep Learning: Understanding of neural network architectures and applications;

• MLOps/LLMOps/AgentOps: Knowledge of production ML infrastructure and DevOps practices for all ML paradigms. Cloud and Infrastructure (AWS Required)

• AWS Expertise: Advanced knowledge of AWS ML and data services (SageMaker, Bedrock, Lambda, ECS, etc.);
• Amazon Bedrock: Deep understanding of Bedrock agents, knowledge bases, and model hosting options;
• Multi-Cloud Awareness: Understanding of Azure, GCP alternatives for comparative discussions;
• Serverless Architectures: Experience with Lambda, API Gateway, Step Functions for agentic workflows;
• Cost Optimization: Ability to design cost-effective solutions with clear TCO analysis;

• Security and Compliance: Understanding of data security, privacy, and compliance requirements. Data Architecture

• Data Pipelines: Understanding of ETL/ELT patterns and tools;
• Data Storage: Knowledge of databases, data lakes, vector databases, and warehouses;
• Data Quality: Understanding of data validation and monitoring;

• Real-time vs Batch: Ability to design for different data processing needs. Nice-to-Have Technical Skills

• AWS Certifications (Solutions Architect Professional, ML Specialty);
• Experience with specific industries (Finance, Healthcare, Retail, etc.);
• Knowledge of AI ethics and responsible AI practices;
• Experience with edge ML and IoT deployments;
• Published thought leadership (blogs, talks, whitepapers);
• Contributions to open-source agent frameworks or MCP servers.