Vetted PyTorch Professionals

“Backend/AI engineer who has shipped production LLM systems end-to-end, including an AI request-routing service (FastAPI + BART MNLI + OpenAI/Gemini) that improved accuracy ~25% after launch via eval-driven prompt/category iteration. Also built an enterprise document intelligence/RAG platform on Azure (Blob/SharePoint/Teams ingestion, OCR/NLP chunking, embeddings in Azure Cognitive Search) with PII guardrails (Presidio), confidence gating, and scalable event-driven pipelines handling millions of documents.”

“Full-stack engineer with healthcare/AI platform experience (Humana), owning an end-to-end high-risk patient prediction feature from React dashboards through FastAPI/TensorFlow real-time inference to AWS EKS operations. Emphasizes production reliability and contract-driven APIs (OpenAPI + generated TS types), plus strong data integration patterns (Kafka, idempotency, DLQs, backfills) in regulated, high-traffic environments.”

“Early-stage engineer with hands-on embedded prototyping experience (Arduino/Raspberry Pi) who helped build an award-winning smart glasses project enabling phone notifications via Bluetooth. Strong computer vision performance optimization background, including accelerating 120 FPS inference by moving from TensorFlow to PyTorch and deploying through ONNX + TensorRT quantization, plus Docker-based GPU deployment and CI/ML practices.”

“Full-stack engineer at an early-stage startup (AirKitchenz) who owned the hourly booking/availability and first paid booking flow end-to-end—React/TypeScript frontend, Node backend, Postgres modeling, and Stripe payments/webhooks. Experienced operating production on AWS (EC2/Elastic Beanstalk, Docker, RDS, CloudWatch) and building reliable, idempotent integrations while iterating quickly in a pre-PMF environment through direct host/renter feedback.”

“Built a production LLM/RAG document extraction and game/quiz content workflow using LLaMA 2, LangChain/LangGraph, and FAISS, achieving ~94% accuracy and reducing turnaround from hours to minutes. Demonstrates strong applied MLOps/orchestration (CI/CD, MLflow, Databricks/PySpark), robust handling of noisy/variable document layouts (layout chunking + OCR fallbacks), and practical reliability practices (human-in-the-loop routing, drift monitoring, A/B testing).”

“Early-career backend engineer and solo builder of FrameFindr, an AI/OCR-based marathon photo tagging product used at live events. Demonstrated pragmatic scaling under tight infrastructure constraints (2GB VPS) and hands-on ownership of architecture, API design, auth (Google OAuth/JWT), and a MongoDB-to-MySQL migration with data-integrity safeguards.”

“Stony Brook Fall 2024 capstone contributor who built a ROS2-based warehouse mobile robot prototype, owning perception and SLAM integration end-to-end. Strong in real-time robotics optimization on Jetson Orin (TensorRT/CUDA, ROS2 tracing/Nsight) and in distributed ROS2 communications (DDS discovery/QoS, MAVLink-to-ROS2 bridging), with a full simulation/testing/deployment toolchain (Gazebo, CI tests, Docker/K3s).”

“Backend/AI engineer with internship experience building an AI-powered financial insights platform (FastAPI, Redis, BigQuery) and prior HCL experience leading a monolith-to-microservices refactor (Flask, Kafka) using blue-green deployments. Demonstrates strong performance/security focus (OAuth/JWT/RBAC, encryption) and measurable impact on latency, downtime, and ML model reliability; MVP was submitted to Google’s accelerator program.”

“AI/LLM engineer who has shipped production RAG chatbots using LangChain/OpenAI with FAISS and FastAPI, focusing on real-world constraints like context windows, concurrency, and latency (reported ~40% latency reduction and <2s average response). Experienced orchestrating AI pipelines with Celery and fault-tolerant long-running workflows with Temporal, and has applied NLP model tradeoff testing (Word2Vec vs BERT) to drive measurable accuracy gains.”

“Built and shipped an agentic RAG chatbot module for NexaCLM to answer questions across large volumes of contracts while minimizing hallucinations and incorrect legal interpretations. Implemented routing between vector retrieval and ReAct-style agent retrieval plus an automated grading/validation layer (cosine-similarity thresholds, retries) and deployed via GitHub Actions to Azure Container Apps, partnering closely with legal stakeholders to define risk/clause-focused objectives.”

“Built and productionized an agentic RAG assistant using Ollama + LangChain + MCP + ChromaDB to speed up and standardize access to operational knowledge from tickets and runbooks. Focused on real-world reliability: mitigated timeouts/latency with retries and concurrency limits, improved retrieval via chunking/embedding iteration, and reduced hallucinations through citation-grounding and confidence-based abstention. Also partnered with non-technical ops staff to deliver anomaly detection/monitoring by translating operational needs into model signals, thresholds, and alerting logic.”

“LLM engineer who has deployed a production system combining LLMs with causal inference (DoWhy) to enable counterfactual “what-if” analysis for experimental research, including a robust variable-mapping/validation layer to reduce hallucinations. Also partnered with non-technical operations leadership at Irriion Technologies to deliver an AI-assisted onboarding workflow that cut onboarding time by 50% and reduced manual errors by ~40%.”

“AI/LLM engineer who has shipped a production RAG platform at Ticker Inc. on GCP (Qdrant + Postgres) delivering sub-second retrieval over 550k+ items, with measurable gains in latency and answer quality (HNSW optimization, MMR re-ranking). Also built an asynchronous LangChain/LangGraph multi-agent research system (10x faster cycles) and partnered with Indiana University doctors on synthetic patient records and ML error analysis using clinician-friendly F1/loss dashboards.”

“AI/LLM engineer with production experience building an LLM-powered investment recommendation system using RAG and chatbots, deployed via Docker/CI/CD and scaled on Kubernetes. Demonstrated measurable performance wins (sub-200ms latency) through QLoRA fine-tuning and TensorRT INT8/INT4 quantization, plus strong MLOps/orchestration background (Airflow ETL + scoring, MLflow monitoring) and stakeholder-facing delivery using demos and Tableau dashboards.”

“Designed and deployed end-to-end LLM-powered AI agent systems to automate knowledge-intensive workflows across marketing/GTM, recruiting, and support. Brings production reliability rigor (evaluation pipelines, monitoring, testing, A/B experiments) plus orchestration expertise (Airflow, Prefect, custom Python) and a track record of translating non-technical stakeholder goals into working AI solutions (e.g., personalized customer engagement agent at Lara Design).”

“ML/computer-vision practitioner who shipped a CycleGAN-based bilingual handwriting translation demo (English↔Telugu) for low-resource scripts using unpaired datasets, focusing on preserving handwriting style and real-time deployment via Gradio. Also delivered a medical imaging pipeline by fine-tuning ResNet-50 and ViT-B/16 for pneumonia detection, emphasizing reproducibility, measurable evaluation, and stakeholder-friendly iteration.”

“Python backend engineer (BetterWorld Technology) who owns microservice systems end-to-end on Azure, including Kubernetes deployments, CI/CD, and production monitoring/alerting. Has hands-on experience integrating SQL/NoSQL (including Cosmos DB with vector search/graph workflow) and has built a Kafka + Spark Streaming pipeline to Snowflake with a reported 40% latency reduction.”

“AI/LLM engineer who built a production resume-parsing and candidate-matching platform at Quadrant Technologies, combining agentic LangChain workflows, VLM-based document template extraction (~85% accuracy), and a hybrid RAG backend for resume-to-JD search. Notably integrated automated LLM evals and metric-based CI/CD quality gates to catch silent prompt/model regressions, and led a 3-person team across frontend/backend/testing.”

“Built a production-style LLM document summarization/generation workflow that mitigates token limits and reduces hallucinations using semantic chunking, FAISS-based embedding retrieval (top-k via cosine similarity), and section-wise generation. Orchestrated the end-to-end pipeline with AWS Step Functions and aligned outputs with sales stakeholders through demos, visuals, and documentation.”

“Software engineer experienced delivering customer-facing, real-time industrial monitoring dashboards (motors/shafts/turbines) by partnering directly with end users to refine charts, alerts, and performance. Strong in API/platform integrations and production troubleshooting—uses feature flags, logging, validation/mapping, containerization, and performance testing to keep systems stable while iterating quickly.”

“Robotics software engineer focused on perception/SLAM and systems integration, recently built a quasi-dynamic mapping pipeline to track and reconstruct articulated objects (e.g., drawers) from RGB video using SAM2, COLMAP SfM, and 3D Gaussian Splatting. Also has strong ROS2 sensor-pipeline experience (custom messages, MCAP rosbag deserialization, tf2) and demonstrated real-time performance tuning by accelerating an ICP-based LiDAR SLAM component ~30x (from ~3s to <100ms per frame).”

“LLM/AI engineer who built a production automated document-understanding pipeline on Azure using a grounded RAG layer, designed to reduce manual review time for unstructured financial documents. Demonstrates strong real-world scaling and reliability practices (Service Bus queueing, Kubernetes autoscaling, observability, retries/circuit breakers) plus rigorous evaluation (shadow testing, replaying traffic, multilingual edge-case suites) and stakeholder-friendly, evidence-based explainability.”

“LLM/RAG practitioner who has taken a support-ticket triage automation system from prototype to production, building the full pipeline (fine-tuned models, FastAPI inference services, vector storage, monitoring) and delivering measurable impact (~40% reduction in triage time). Demonstrates strong operational troubleshooting of LLM/agentic workflows (observability-driven debugging, fixing agent routing/looping) and supports adoption through tailored demos and sales-aligned technical communication.”