AI Software Engineer - Senior

FSS Government Solutions is seeking a Senior Python Developer to design and deliver cloud\-native, event\-driven backend platforms across **Microsoft Azure and Amazon Web Services**, including platforms that host and operate AI\-enabled products. You will own architecture and implementation of high\-reliability workflows, integrate internal and external systems at scale, distribute metadata and events under SLA, and optimize service design for LLM access, retrieval, cost, latency, and reliability. You will provide technical leadership on production Python services and on the interfaces that connect those services to models and agent tooling in a dual\-cloud environment. * Role is 100% Remote * Candidates must obtain a Public Trust Clearance **What you will do** * Lead design and implementation of cloud\-native, event\-driven platforms that integrate many independent systems into a unified operational ecosystem spanning Azure and AWS. * Architect durable, event\-driven workflows using Azure Durable Functions / Service Bus / Event Grid **and** AWS Step Functions / EventBridge / SQS / SNS to orchestrate high daily event volumes with reliability, retries, and recovery. * Build Python backend services that generate and distribute structured metadata and events to downstream consumers under tight availability SLAs, including near\-real\-time propagation targets, regardless of which cloud owns a given path. * Design scalable, fault\-tolerant microservice\-style components that can be extended or moved between clouds without rewriting the platform. * Integrate internal and third\-party systems through REST and GraphQL; use Azure API Management and/or Amazon API Gateway as governed gateways. * Design and implement AI product architecture on both clouds: model\-serving and inference paths (Azure OpenAI / Azure AI Foundry and Amazon Bedrock), retrieval / grounding pipelines, prompt\-and\-tool orchestration, and secure service interfaces for LLM interaction. * Build MCP\-style and function\-calling interfaces so agents and LLMs can invoke platform APIs safely (auth, scoped tools, timeouts, audit). * Optimize AI workloads for production: token and compute cost, latency budgets, rate limits, caching, fallback models across providers, and evaluation / tracing of model calls. * Implement durable storage with Azure Blob / Table Storage and Amazon S3 / DynamoDB for event\-driven exchange, recovery, embeddings or document corpora, and audit of AI and integration traffic. * Establish observability across application and AI paths: structured logging, distributed tracing, prompt/tool telemetry, and alerting in Application Insights **and** CloudWatch / X\-Ray (or an equivalent unified telemetry layer). * Define cloud\-portable patterns (interfaces, adapters, IaC) so services are not hard\-wired to a single vendor’s SDK. * Influence architecture, raise code quality, and mentor engineers on backend, dual\-cloud, and AI\-integration patterns. * Translate stakeholder re