Vetted PyTorch Professionals

Ahmed Sadaqat

Screened

Senior Machine Learning Engineer specializing in production ML and predictive analytics

Los Angeles, CA7y exp

“ML/AI engineering leader who has owned end-to-end production systems from experimentation through deployment, monitoring, and iteration at meaningful scale. They describe running a 1M+ records/day prediction platform with 99.9% availability, shipping a RAG-based conversational AI feature for 50,000 active users, and consistently improving precision, latency, reliability, and cost with measurable business impact.”

“Backend/data engineer from Oscar Health specializing in healthcare claims systems on AWS. Built HIPAA-compliant real-time services (FastAPI/Postgres/Kafka on EKS) and serverless ingestion pipelines, and led modernization of a legacy SAS claims pricing system to Python/Spark with rigorous parity validation. Demonstrated measurable impact with high uptime/low latency services and major Snowflake performance and cost reductions.”

“Built and shipped a production LLM-powered medical scribe that generates structured clinical visit summaries using RAG, strict JSON schemas, and post-generation validation to reduce hallucinations. Experienced in making LLM workflows deterministic and observable (structured logging/metrics/tracing) and in evaluation-driven iteration with metrics like schema pass rate and edit rate; collaborated closely with clinicians and policy stakeholders at Scale AI to drive adoption.”

“Built and shipped a real-time AI phone agent for small businesses that handles bookings/FAQs/messages using streaming ASR, an LLM with tool-calling, and TTS; deployed to production for multiple paying customers. Demonstrates strong applied LLM reliability practices (tool-first grounding, retrieval, hard-negative testing, and production monitoring) and experience orchestrating multi-step AI workflows with Airflow, Prefect, and AWS Step Functions.”

“Frontend engineer from Amazon Advertising who owned a sophisticated React/TypeScript ad creative builder used by advertisers and ad ops teams. Stands out for combining deep browser-level debugging with product-minded UX improvements that reduced support escalations and made complex multi-placement ad configuration faster and more reliable for power users.”

“Front-end/product engineer with strong financial-domain experience, including browser-based interfaces for retail planning, sales forecasting, and high-throughput trading data. Stands out for combining UI engineering with machine learning, time-series prediction, browser performance optimization, and functional programming across tools like React, TypeScript, Elm, OCaml, and Scala.”

“Robotics/embedded engineer with hands-on experience building real-time control systems on RP2040 (hydroponics automation, 1-DOF helicopter stabilization) and full ROS 2 navigation stacks in simulation (URDF, TF, PID, A* in RViz/Gazebo). Demonstrates strong low-level protocol work (timing-sensitive one-wire in C) and rigorous debugging across hardware and software using UART instrumentation and oscilloscope verification, plus reproducible workflows with Docker and CI/CD (GitHub Actions/GitLab, incl. Sandia National Labs).”

“Robotics engineer who helped bootstrap Meta’s humanoid robotics effort, building simulation training and deployment infrastructure for vision-language-action (VLA) models. Evaluated multiple physics backends (Bullet, MuJoCo, Isaac, internal) to minimize sim-to-real gap and addressed control-loop frequency mismatches via sequence optimization/MPC-like approaches and trajectory-output modifications. Published research that contributed a new addition to ROS 2 and has built ROS2 node stacks spanning control, perception, teleop, tactile sensing, and imaging.”

“Backend/AI engineer who built a production GPU-backed real-time inference API at Nvidia and debugged burst-induced tail latency, cutting P95 by ~29% through dynamic batching and backpressure. Also shipped an end-to-end RAG + agentic operational diagnostics assistant with strict tool controls, evidence citation, confidence gating, and strong production guardrails, plus demonstrated hands-on Postgres optimization (900ms to 40–60ms).”

David Gross

Screened

Entry AI Software Engineer specializing in LLM workflows and ML pipelines

Redmond, WA2y exp

“Built an autonomous-agent document indexing concept in a hackathon with Microsoft and The Seattle Times, architecting an Azure-based system (Azure AI Foundry, Cosmos DB, Azure indexing, Copilot Studio) and coordinating closely with the customer team. Also created and pitched a sports matchmaking app (Ludicon), combining user studies, feature implementation, and technical support on sales/investor calls.”

Aditya Sawant

Screened

Mid-level Software Development Engineer specializing in robotics and cloud-based device management

North Reading, Massachusetts3y exp

“Amazon Robotics engineer who deployed and scaled the Lumos camera-based package scanning work cell across EU sort centers (100+ work cells in 5+ sites), enabling remote launches via detailed runbooks and troubleshooting. Strong in AWS IoT/edge systems, with hands-on incident recovery (restored 34 down work cells) and secure multi-compute certificate provisioning using IoT Jobs, ACM/CA, and custom roles; delivered ~75% per-cell cost reduction vs Cognex-based approach.”

Pankaj Goyal

Screened

Director-level Engineering Leader specializing in FinTech, IAM, and AI/ML platforms

SF Bay Area, CA22y exp

“Player-coach backend leader at PostLo who led a major backend architecture upgrade to enable AI-driven features by separating transactional systems from AI workloads (vector embeddings/image validation) and adding async processing for heavy jobs. Also owned production reliability improvements (query/index optimization, workload isolation, monitoring and load testing) and translated an ambiguous retention goal into a shipped cashback rewards feature with auditable transactions.”

“Long-tenured Intel engineering leader with 27 years of experience spanning hardware-near software through cloud and web layers, including Chromebook and AI platform work. Particularly compelling for senior platform or AI infrastructure roles: he led cross-functional efforts to unify Intel's oneAPI stack across PC, embedded, and data center environments and helped drive LLM inference performance from roughly 11 to 22 tokens per second.”

“CMU engineer with hands-on experience building full-stack AI products across internships and academic projects, including a Microsoft real-time multimodal assistant and a Cursor-like coding agent. Stands out for combining low-latency systems engineering with strong product instincts, shipping AI tools that improved latency by 35%, cut grading time from five days to under 10 minutes, and reduced user resubmissions by 45%.”