Vetted Requirements Gathering Professionals

“Light industrial staffing operations leader who scaled from multi-unit branch manager to Market VP overseeing four Dallas–Fort Worth branches (~$63M annual volume) and 56–58 direct reports. Known for data-driven SOP/operations turnarounds that lifted revenue 30%+ and gross profit 25%+ (President’s Club top 10%), plus leading a multi-market Employee Bridge mobile app pilot and tightening E-Verify compliance for high-volume accounts like HelloFresh.”

“Full-stack engineer who built and scaled an AI-powered web product (React/Next.js + TypeScript/NestJS) with MongoDB, Redis, and RabbitMQ. Strong in rapid iteration while maintaining production quality—uses versioned APIs, feature flags, CI/CD, and observability (correlation IDs/structured logs) to ship frequently and debug distributed workflows. Also created an internal operations dashboard for real-time visibility and control of background jobs/AI workflows that was adopted quickly by ops and product teams.”

“Built and productionized an internal GenAI chatbot that makes company policy/SOP knowledge instantly searchable, using a secure RAG architecture on AWS (Bedrock/Titan embeddings/OpenSearch Serverless, Textract/Lambda/S3 ingestion, Claude 3 Sonnet). Demonstrates strong MLOps/orchestration experience (Airflow, Step Functions with Lambda/Glue/SageMaker) and a rigorous reliability approach (RAGAS metrics, A/B testing, citation validation, monitoring), including collaboration with compliance stakeholders via review dashboards.”

“Product intern at an AI startup (AdvisorGPT) who helped turn an LLM-based prototype into a production SEO blog-generation workflow that matched a firm’s tone/voice and targeted specific search phrases. Strong at bridging technical and non-technical teams, rapidly learning new AI tooling, and driving adoption through customer calls, UX improvements, and customer-facing demos/workshops.”

“Built and shipped a production RAG system (“TROJAN KNOWLEDGE”) for answering questions over technical PDFs, using a 3-stage retrieval stack (BM25 + FAISS + cross-encoder) to lift F1 from 71% to 84%. Drove major performance gains with a 3-level cache (memory/Redis/disk) cutting latency from ~200ms to ~10ms, and added Prometheus/Grafana monitoring plus LangChain-based fallback logic to handle OpenAI rate limits under load.”