Forward Deployed Engineering

About the Company

COMPANY OVERVIEW AND JOB SUMMARYUSTECH Solutions is a global IT services partner to Fortune 500 companies, delivering AI, Data, Cloud and Application Services at enterprise scale. Our work sits inside our clients' core systems — in financial services, insurance, healthcare and enterprise SaaS — where engineering quality, security and regulatory posture matter as much as speed. The Forward Deployed Engineering practice is how we bring that capability directly into a client's engineering organization.

About the Role

The FDE Lead is the senior technologist we embed inside a client's engineering organization to turn an AI ambition into running, supported software. The role carries two jobs at once: solution architect for the application estate being changed, and engineering lead for the agentic systems built on top of it. You own the architecture from first business requirement through production release and hypercare, and you lead the pod that ships it. FDE Leads are embedded from week one and work in small senior pods, helping the client decide what is worth building, building it, and staying accountable through to production support. The role carries unusual latitude: you shape the roadmap alongside client architects, make the technology calls, and answer for the outcome. You are the technical face of USTECH Solutions in front of client engineering leadership and CXOs.

KEY RESPONSIBILITIES: APPLICATION SOLUTION ARCHITECTURE

• Own end-to-end solution architecture for the applications in scope: current-state assessment, target-state blueprint, HLD and LLD, interface contracts and the migration path between them.
• Select the stack and patterns deliberately — microservices vs. modular monolith, event-driven vs. request-response, API gateway, caching, transaction boundaries — with trade-offs documented.
• Design the integration layer to legacy and packaged systems (core banking, ERP, mainframe, CRM) through secure abstraction layers, API facades and event streams rather than point-to-point coupling.
• Define the shared data architecture: canonical models, data contracts, lineage, masking and residency rules.
• Set and defend non-functional requirements — availability, latency, scalability, resilience, DR, observability, cost per transaction — and prove them through load and failure testing.
• Build security and compliance into the design (authN/authZ, secrets, encryption, SOC 2, ISO 27001, GDPR, HIPAA as applicable) and produce the artifacts the client's architecture review board needs: ADRs, design documents, sequence and deployment diagrams, threat models, estimation models.
• Govern engineering quality across the pod: code standards, review discipline, test coverage, CI/CD, IaC and technical debt tracking.

KEY RESPONSIBILITIES: AGENTIC AI AND PLATFORM ENGINEERING

• Architect and deploy intent-driven agentic workflows against real domain logic — underwriting, claims, regulatory checking, reconciliation, servicing, order-to-cash, IT operations.
• Build and tune context-aware RAG pipelines and vector store architectures over disparate enterprise sources, owning chunking, embedding, retrieval evaluation and drift.
• Implement Model Context Protocol (MCP) servers, tool and function calling, and multi-agent orchestration so agents act on enterprise systems within scoped permissions.
• Stand up evaluation frameworks and AI observability: golden datasets, regression suites, accuracy thresholds, latency and token cost tracking, tracing and prompt versioning.
• Establish guardrails and AI governance — deterministic controls, human-in-the-loop checkpoints, PII redaction, audit trails, model and prompt change management.
• Apply AI-native engineering to our own delivery (agent-assisted development, automated test and edge-case generation) and evidence the productivity gained.
• Stay platform-pragmatic across Azure AI Foundry, AWS Bedrock, Google Vertex AI, OpenAI and Anthropic, choosing per problem, cost envelope and client constraint.

KEY RESPONSIBILITIES: CLIENT ENGAGEMENT AND DELIVERY LEADERSHIP

• Embed inside client engineering teams as a trusted technical advisor, operating with high autonomy through discovery ambiguity and directly in front of stakeholders.
• Translate business intent into technical specifications the pod and the agents can execute against, including domain constraints and acceptance criteria; challenge weak use cases early rather than building them.
• Present architecture, security posture, evaluation results and business outcomes to client engineering leadership and C-level stakeholders (CTO, CIO, CPO, Chief Risk Officer).
• Lead a small senior onshore-offshore pod through shaping, build, evaluation, release and hypercare, with visible delivery governance and honest status against committed outcomes.
• Support pre-sales for the practice: solution shaping, RFP and RFI responses, technical proposals, estimation, pricing inputs, proofs of concept and client workshops.
• Mentor engineers and architects, and contribute reusable accelerators, reference architectures and playbooks to the USTECH Solutions asset library from every engagement.

KEY MEASURES OF SUCCESS

• Production deploymentsEvery engagement reaches production; no prototype handed over without a production path
• Client outcomeValue evidenced in the client's own metrics, and a client willing to act as a reference
• Engineering velocityMeasured reduction in feature-to-production cycle time, review turnaround and manual test effort
• Quality and trustEvaluation and accuracy thresholds met before release; no security or compliance escalations
• Account growthFollow-on scope originating from the embedded work
• Reusable IP and talentAccelerators contributed from every engagement; pod members taking on architecture ownership

MUST-HAVE QUALIFICATIONS AND EXPERIENCE

• 15+ years in software engineering, application architecture, platform engineering or solution architecture, including 8+ years delivering high-scale enterprise products or services.
• Proven solution architect track record on the application side: owning target-state design, integration architecture, NFRs and technical estimation for multi-team programs.
• Full-stack depth across at least one primary ecosystem (Java / Spring, .NET, Node or Python), plus APIs, messaging, relational and NoSQL data stores, and cloud-native deployment on Azure, AWS or GCP.
• 3+ years hands-on with the modern AI stack — RAG, vector databases, prompt engineering, tool and