Vetted PyTorch Professionals

“Built and productionized an LLM-driven document verification workflow for a construction firm’s submittals process, moving from a Vercel/Next.js prototype to a FastAPI + LangChain/LangGraph backend with background workers and multi-server deployment. Uses LLM tools (e.g., OpenAI Codex/Cloud Code) for rapid development and log-driven root cause analysis, and partners with customer teams on evaluation metrics and iterative improvements.”

“Robotics/AI Master's thesis researcher building an LLM-driven workflow to generate and evaluate robot policies before running them in an environment. Also built a local LLM-based real-time target-tracking robot using a pan-tilt camera with LangChain + Ollama, and has hands-on ROS 2/Gazebo experience including URDF-based simulation and a TurtleBot multi-agent chase project.”

“Built a production LLM-powered fraud detection platform at Wells Fargo, combining OpenAI/Hugging Face models with RAG-based explanations to make flagged transactions interpretable for risk and compliance teams. Delivered low-latency, real-time inference at high scale on AWS (SageMaker + EKS), with strong observability and security controls, reducing manual reviews and false positives in a regulated environment.”

“Built and shipped a production AI-powered Q&A/RAG onboarding assistant at One Community Global that unified knowledge across Notion, Google Docs, and Slack, cutting volunteer onboarding time by 45%. Demonstrates strong end-to-end ownership: LangChain agent orchestration integrated into a FastAPI backend, rigorous evaluation (200-query dataset, ~85% accuracy), and production feedback/monitoring with source-attributed answers to build user trust.”

“Internship experience building an end-to-end medical AI pipeline that extracts and normalizes messy medical PDFs, fine-tunes BioBERT to classify tumor-related statements (including negation/ambiguity handling), and integrates image-model outputs (MedSAM/GroundingDINO) for tumor localization and classification. Also worked on an LLM/RAG system to draft IPO prospectuses using retrieved regulatory/financial sources (including SEC EDGAR) with structured prompts to reduce hallucinations.”

“Backend/ML engineer who has shipped high-scale real-time systems across e-commerce and healthcare: built a PharmEasy real-time recommendation engine for ~2M monthly users (cut feature latency 5 min→30 sec; +15% cross-sell) and architected a HIPAA-compliant multimodal clinical diagnostic workflow (DICOM+EHR) with XAI, MLOps (MLflow/Airflow/K8s), and drift/monitoring guardrails supporting 10k+ daily predictions.”

“Product-focused full-stack engineer at startup (Zippy) who shipped a production multi-agent AI system for restaurant operations plus payments workflows. Built end-to-end: RAG grounded on a Notion knowledge base, structured function-calling task routing, FastAPI/JWT multi-tenant backend, and a polished React+TypeScript owner dashboard. Has real production incident experience (duplicate Stripe webhooks) and reports ~94% task-routing accuracy under load.”

“Interned at Larsen & Toubro where they built and deployed an agentic RAG document question-answering system to reduce time spent searching documents and improve trustworthiness. Implemented ReAct-style multi-step orchestration with LangChain/LlamaIndex plus evidence-bounded generation, grounding/citations, and rigorous evaluation—cutting latency ~40%, hallucinations ~35%, and unsafe outputs ~40% while collaborating closely with non-technical business/ops stakeholders.”

“Robotics software engineer/researcher focused on perception, SLAM, and sensor fusion, with hands-on experience taking systems from simulation to embedded/real-time deployment. Led transparent-surface (glass) detection using GDNet and achieved a major real-time speedup (~7–9 FPS to ~30 FPS) while preserving >90% recall, and has built ROS-based EKF GPS-IMU fusion plus profiled/optimized Visual SLAM for performance and memory stability. Also brings production-style deployment skills via Docker/Kubernetes orchestration of ML inference services with autoscaling and model update rollouts.”

“AI engineer who has shipped production LLM systems end-to-end, including a natural-language-to-SQL analytics copilot for career advisors that achieved ~95% query success through schema grounding, access controls, and automated regression testing with golden queries. Also builds LangGraph-orchestrated multi-step agents (resume analysis, recommendations) and RAG pipelines (PDF ingestion + FAISS) and partners closely with non-technical users to drive adoption and trust.”

“Built and deployed an LLM/RAG system that uses differential privacy and distributional similarity checks to transform private data into a non-sensitive knowledge base while preserving utility. Also has experience demonstrating adversarial ML concepts (FGSM) to non-technical audiences by focusing on observable model behavior rather than implementation details.”

“Built and deployed production LLM pipelines for evidence-based scoring in two domains: biomedical literature mining (scoring ~2700 drug compounds vs gene targets/mechanisms) and long-horizon news analytics (35 years of Chinese articles). Emphasizes reliability at scale (retries/checkpointing/validation), rigorous empirical model benchmarking (GPT-4o/mini/5), and translating results into stakeholder-friendly visual narratives.”

“ML/AI practitioner in healthcare (Syneos Health) who has deployed production clinical NLP and risk models. Built a BERT-based physician-note information extraction system on Docker + AWS SageMaker (reported ~42% retrieval improvement) and automated retraining/deployment with Airflow and drift detection, while partnering closely with clinicians to drive adoption (reported ~18% readmission reduction).”

“Master’s student and Deep Learning teaching assistant who teaches LLM/VLM fine-tuning (including LoRA) and built a Hugging Face LLM fine-tuned for unit conversion, improving reliability by analyzing synthetic data and filling missing number-system conversion examples. Also implemented the Raft consensus protocol using gRPC in a distributed systems course with correctness validated by unit tests.”

“Built and deployed a GPT-4-powered medical assistant for clinical staff to reduce time spent searching guidelines and EHR information, with a strong emphasis on safety and compliance. Uses strict RAG, confidence thresholds, and fallback behaviors to prevent hallucinations, and runs production-grade workflows orchestrated with LangChain/LangGraph plus Docker/Kubernetes/MLflow and monitoring for reliability and cost.”

“Built a production multi-cloud LLM-driven IT ticket automation system using LangGraph, Azure + Pinecone RAG, and an Ollama-hosted LLM on AWS, with Terraform-managed infra and PostgreSQL audit/state tracking for reliability. Also partnered with UW School of Medicine & Public Health students to deliver a glioma survival risk-ranking model, translating clinical feedback into practical pipeline improvements (imputation, site harmonization) and stakeholder-friendly visualizations.”

“AI/ML infrastructure-focused engineer who has built production RAG systems from scratch (Supabase/pgvector + OpenAI embeddings) and iterated using formal eval metrics to improve retrieval quality. Also debugged real-time audio issues in a LiveKit-based pipeline by correlating packet loss with VAD behavior, and has deep experience building brittle, customer-specific financial platform integrations in Python/Playwright (2FA, redirects, token refresh, rate limits).”

“LLM/ML engineer who has shipped an enterprise RAG-based Q&A system (LangChain/LlamaIndex, FAISS + Azure Cognitive Search, GPT-3.5/4 via OpenAI/Azure OpenAI) to production on Docker + Kubernetes/OpenShift, tackling hallucinations, retrieval quality, latency/cost, and RBAC/IAM security. Also partnered with operations leaders to turn manual reporting into an LLM-powered summarization and forecasting dashboard driven by real KPIs and iterative stakeholder feedback.”

“Built and deployed a production LLM-powered document Q&A system enabling natural-language querying of large PDFs, focusing on retrieval quality (overlapped chunking) and low-latency performance (optimized embeddings + vector search). Experienced with scaling ML/LLM workflows using async/batch processing, caching, cloud storage, and orchestration via Apache Airflow with robust testing, monitoring, and failure handling.”

“LLM/agentic workflow specialist with healthcare deployment experience who has taken LLM-based automation from prototype to production using operator-in-the-loop validation, RAG-style retrieval, RBAC, and monitoring for sensitive data compliance. Demonstrated real-time incident resolution (retrieval timeouts due to network/proxy misconfig) and strong GTM support—hands-on developer workshops and sales demos translating technical safeguards and real-time ETL into measurable ROI (70% ops reduction, ~$200K/year savings).”

“Built and shipped multiple production LLM/agentic systems, including an agentic RAG NL-to-SQL analytics app that cut manual reporting from 9 hours/week to 15 minutes by grounding on schema-aware retrieval and robust fallback/monitoring. Also implemented a LangChain supervisor-orchestrated enterprise IT automation agent that routes requests for search, identity validation, and action execution, and created a RAG search tool spanning Jira/Confluence/SharePoint for operations stakeholders.”

“Built a production real-time anomaly detection platform for high-frequency trading at HSBC, using a streaming stack (Pulsar + Spark Structured Streaming + AWS Lambda) and a transformer-based model combining time-series and numerical signals. Experienced in MLOps and safe deployment (Kubernetes, canary releases, MLflow/Grafana monitoring) and in aligning model performance with risk/compliance expectations through SLA-driven tuning and stakeholder-friendly dashboards.”

“Healthcare-focused applied ML/LLM engineer who has deployed production systems including an LLM medical documentation assistant that summarizes unstructured EHR notes into physician-ready structured outputs. Experienced building secure, compliant pipelines (PHI minimization, RBAC, encryption) and scaling via Docker/Kubernetes/Azure ML, plus orchestrating ETL/ML workflows with Airflow and Kubeflow; also built an LLM-driven clinical coding assistant at Centene with measurable performance metrics.”