Vetted Model Deployment Professionals

“Built and deployed a production LLM-powered document Q&A system using a strict RAG pipeline (LangChain-style orchestration + FAISS) to help users query large internal document sets. Demonstrates strong reliability focus through hallucination mitigation, curated offline evaluation with grounding checks, and production monitoring (latency/fallback rates) plus stakeholder alignment via demos and business metrics.”

“AI/back-end engineer at Canon who helped build and operate an internal production LLM platform that acts as a secure middle layer between users and models, defending against jailbreaks/prompt injection while enabling RAG, memory, and grounded responses over company data. Experienced with LangChain/LangGraph orchestration, vector DB retrieval, and reliability practices (testing, monitoring, adversarial prompts) to run high-throughput, low-latency AI workflows in production.”

“ML/NLP practitioner who has delivered production systems in regulated domains, including a healthcare compliance pipeline using RAG (GPT-4/Claude) plus TF-IDF retrieval that increased document review throughput 4.5x. Also has hands-on experience improving fraud detection data quality via entity resolution (Levenshtein, Dedupe.py) validated with A/B testing, and building scalable, monitored workflows with Airflow, CI/CD, and AWS SageMaker.”

“Unity gameplay engineer with hands-on ownership across core game systems, multiplayer, AI/LLM integration, and cross-platform XR/mobile shipping. Particularly compelling is their combination of deep runtime optimization and systems architecture: they cut match-3 cascade frame spikes by 40-50%, eliminated a class of production race-condition bugs, and also built a local-model FastAPI-backed AI prototype with structured evaluation and cost control.”

“Built and shipped a production Python-based agentic RAG document retrieval system over 80K records using FastAPI, OCR, vector search, and AWS infrastructure, with a strong emphasis on reliability, testing, and observability. Stands out for treating AI failures like production incidents—turning hallucinations, retrieval misses, and OCR issues into regression tests—and for quantifiably reducing document lookup time from about 12 minutes to under 90 seconds.”

“Backend engineer who rebuilt and refactored high-traffic systems at Phenom using Java/Spring Boot/Play and also designs Python/FastAPI services. Focused on measurable reliability and performance gains through DB/query optimization, async processing, and strong observability, with disciplined rollout practices (feature flags, parallel runs, rollback) and security patterns including token auth and row-level security.”

“Built and deployed a production LLM-assisted recommendation and insights platform that unifies structured, semi-structured, and unstructured data via a modular ingestion pipeline, canonical schemas, embeddings, and late-fusion modeling. Experienced in operationalizing ML/LLM systems with Airflow and Kubernetes (Dockerized services, autoscaling, rolling updates) and emphasizes reliability through layered testing, guardrails, monitoring, and A/B experimentation while partnering closely with non-technical stakeholders.”

Siddhartha Geddam

Screened

Junior Robotics/ML Engineer specializing in autonomous UAVs and perception

2y exp

“Machine learning robotics engineer with internship experience deploying object detection and semantic segmentation models to an autonomous vehicle fleet operating in airports and naval docking stations, optimizing with ONNX/TensorRT for NVIDIA Jetson edge deployment. Also built ROS/ROS2-based decentralized multi-drone coordination (TF trees, shared telemetry) validated in Gazebo and networked via Nimbro with sub-10ms latency messaging.”

“Built a production ad-spend optimization system that combined deterministic audit logic with LLM-generated explanations, surfacing severe inefficiencies including 70-90% wasted spend in some Google Ads accounts. Stands out for pairing measurable business impact with pragmatic AI safety and usability decisions, including approval-gated execution and structured, human-readable recommendations.”

Siddharth Ranjan

Screened

Mid AI/ML Engineer specializing in LLMs, RAG, and cloud AI systems

Maryland, USA4y exp

“Built an AI-powered job matching platform end to end using AWS, Gemini, FastAPI, TypeScript, embeddings, and vector search. The standout result was automating manual matching workflows and scaling resume processing to roughly 2,000 resumes per minute while monitoring quality with F1 score and latency metrics.”