Vetted GPT-4 Professionals

“Built and deployed a production LLM-powered risk assistant for KPMG and Freddie Mac that lets analysts query a confidential Neo4j risk graph in natural language (no Cypher), turning multi-day analysis into minutes with traceable, cited answers. Implemented rigorous guardrails, deterministic verification, RBAC/security controls, and a full eval/observability stack, cutting query error rate by ~50% and iterating through weekly UAT with non-technical risk analysts.”

“Built GenMedX, a multi-module clinical AI system for emergency department decision support spanning triage prediction, diagnosis, medication Q&A, and visit summarization. Stands out for combining medical LLM fine-tuning, RAG, and rigorous evaluation/monitoring to drive a major triage recall improvement from 38.5% to 76.6%, with a strong focus on safety, edge-case detection, and production reliability.”

“Built a production RAG-based natural-language-to-SQL system at Global Atlantic to replace slow, expensive manual analytics ticket workflows, focusing heavily on retrieval quality and measurable evaluation (200-question ground-truth set; recall@5 improved 0.65→0.78 via semantic chunking). Also built a custom MCP-style agent orchestrator for a personal project (arxiv-ai) to improve flexibility and Langfuse-aligned observability, and has hands-on experience with LangGraph, CrewAI, and n8n.”

Ranjani Salla

Screened

Mid-level AI/ML Engineer specializing in LLMs, FinTech, and Healthcare IT

USA5y exp

“Built production GenAI systems in both healthcare and financial services, including a Verily clinical platform and an Accenture financial Q&A product. Stands out for combining advanced RAG, fine-tuning, safety evaluation, and infrastructure engineering to deliver measurable gains in engagement, groundedness, hallucination reduction, and cost efficiency.”

“Built production LLM systems focused on reliability and safety, including a plain-English deployment tool that generates validated plans and provisions to Kubernetes while preventing unsafe actions via schema enforcement and plan/execute separation. Also created multi-LLM workflows (LangGraph) and stakeholder-friendly demos at Bosch, including a PyQt/FastAPI/CUDA app comparing SAM2 vs SAMWISE for on-device object detection with intuitive UX for business users.”

“ML/LLM engineer with production experience building a multi-GPU LLM inference platform using TensorRT and vLLM, achieving ~40% p95 latency reduction through batching/KV caching, quantization, and CUDA/runtime tuning. Also has end-to-end orchestration experience (Kubernetes, Airflow) and has delivered real-time fraud detection systems at Accenture in close collaboration with non-technical risk and product stakeholders.”

“Backend engineer with end-to-end ownership of scientific and AI-powered systems, including neuron imaging pipelines at Monell Chemical Senses Center and an LLM-based structured information extraction platform for Wharton and PSG. Stands out for turning messy, compute-heavy workflows into reliable production backends with measurable impact, including saving researchers over 50 hours per week.”

“Built and deployed PIKA, an internal multi-agent platform for FDA-regulated software development, owning it from concept through production. The candidate combines strong full-stack engineering with rigorous LLM orchestration, human-in-the-loop controls, and production eval systems, delivering measurable impact: 3x more design issues caught, ~90% fewer false positives, and ~40% efficiency gains on documentation-heavy workflows.”

“Built and owned an LLM support copilot at Stripe focused on improving agent ticket resolution. Designed the backend and ML system end to end, using RAG, Redis caching, hybrid vector search, and LoRA fine-tuning to achieve 40% lower latency and 22% higher response accuracy, with continuous quality monitoring via Ragas and related evaluation frameworks.”