Vetted Model Evaluation Professionals

“Built and shipped "OnCall Copilot," a production Slack-based RAG assistant that answers on-call questions from runbooks and postmortems with citations using a FAISS vector index. Emphasizes reliability and measurable performance via strict guardrails ("no evidence, no answer"), evaluation metrics, drift monitoring, and operational hardening with Docker, logging, health checks, and offline fallback.”

“Built and deployed a production AI accessibility platform that turns chart and image-based graphs into real-time audio narratives for visually impaired users. Implemented a ResNet-based CV + OCR + NLP + TTS pipeline and improved performance through preprocessing, Redis caching, and Kubernetes autoscaling/rolling updates on AWS to handle traffic spikes with no downtime.”

“Backend engineer with experience modernizing legacy systems and building modular Python/Flask services, including a REST-to-GraphQL migration for an e-commerce platform that improved API response time by 45%. Strong in performance and scalability work across PostgreSQL/SQLAlchemy (indexing, JSONB, N+1 fixes, connection pooling) and high-throughput systems (Celery + Redis), plus integrating ML microservices with TorchServe, Kafka streaming, feature stores, and Prometheus/Grafana monitoring.”

“Built and deployed a production-grade LLM agent for credit management and accounts receivable automation, integrating ERP/MySQL data via a RAG pipeline and exposing services through FastAPI with Pydantic-validated outputs on AWS Bedrock. Emphasizes reliability and compliance for financial operations using schema validation and human-in-the-loop review, reporting ~32% reduction in manual work and ~41% improvement in response time/reliability.”

“Built production LLM/agent systems for procurement and contract spend controls, including a proactive contract value leakage detection platform that moved an organization from reactive audits to pre-payment rejection. Combines multi-agent orchestration (Semantic Kernel/LangChain/AutoGen), document AI benchmarking (Textract vs Azure DI), and MLOps/testing (MLflow, QTest/Pytest) with strong security practices (RAG-grounded responses to prevent prompt injection). Integrated anomaly alerts directly into SAP SES workflows and Power BI dashboards, citing ~$38M leakage addressed across large spend environments.”

“Frontend engineer with experience across university and product companies (University of Montreal, Dopely, Takhfifan), owning React/TypeScript features end-to-end. Notably built a mathematically complex, multi-mode color wheel UI for designers and led quality practices at scale via conventions, RTL testing, and code reviews for junior developers, plus performance and reusability improvements in existing codebases.”

“Built and deployed a production AI-powered personalized learning platform (Django + FastAPI) featuring an LLM+RAG tutoring assistant and automated grading. Demonstrates strong applied LLM reliability engineering (structured JSON outputs with Pydantic validation, hallucination control via FAISS-based RAG thresholds and refusals) plus scalable async microservice design and Airflow-orchestrated ETL across AWS/GCP.”

“Built a production-style AI meeting summarization and action-item extraction system (Azure Speech-to-Text + transformer summarization/NER) exposed via a Flask REST API, with explicit guardrails to prevent hallucinated tasks. Strong focus on reliability: modular agent/workflow design, precision-first evaluation with human-validated golden notes, and practical orchestration patterns (tool-augmented agents; ready to scale into Airflow/LangGraph/Prefect).”

“Built a production AI-powered university marking system that automates question generation and grading from PDF course materials using a RAG pipeline (S3 + Pinecone) orchestrated with LangChain/LangGraph and deployed on AWS ECS via Docker/ECR and GitHub Actions CI/CD. Addressed a key real-world LLM challenge—grading consistency—by implementing rubric-based scoring, retrieval re-ranking, and standardized context summarization, validated against human instructors.”

“Robotics software candidate who built a Q-learning smart delivery drone navigation system, focusing on 2D path planning with dynamic obstacle avoidance using reward shaping and real-time sensor feedback. Actively learning ROS 2 by building Python simulation projects with publisher/subscriber patterns and has experience coordinating multi-agent drone simulations via message passing.”