Vetted AI & Machine Learning Professionals

“Built and shipped a production LLM hallucination detection and monitoring pipeline using semantic-level entropy (embedding-clustered multi-generation variance) to flag unreliable outputs in downstream automation. Implemented a scalable async architecture (FastAPI + Docker + Redis/Celery) with strong observability (structured logs + PostgreSQL) and developed evaluation loops combining controlled prompts and human review; also partnered with non-technical stakeholders on AI-driven form validation/document processing.”

“ML-focused builder who has shipped an end-to-end income-class prediction product: built the data pipeline, trained models, deployed via Streamlit with a live UI, and tracked success via accuracy (84%), adoption, and latency. Demonstrates strong practical MLOps instincts (Docker/Streamlit Cloud, logging/monitoring, caching) and data engineering reliability patterns (schema checks, idempotency, retries, backfills) while iterating quickly in ambiguous, solo-project environments.”

“AI engineer who built an agentic, production-deployed LLM workflow for tobacco violation parsing and automated multi-case creation, using six specialized agents and a human-in-the-loop confidence-threshold routing design. Addressed data privacy constraints by generating synthetic datasets with LLM prompting, and orchestrated reproducible end-to-end pipelines in LangChain with robust testing and evaluation (precision/recall, micro-F1).”

“Backend/platform engineer who built an LLM-driven QA automation system (“mockmouse”) using a Flask orchestration microservice, Socket.IO real-time updates, Redis caching, and strict Pydantic schemas to turn prompts into reliable action graphs and automated browser tests. Has hands-on Kubernetes delivery experience (Docker/Helm/Jenkins) and has supported large migration programs, validating ETL cutovers with 1M+ synthetic records and rigorous output comparisons; also built event-driven monitoring/anomaly detection streaming into Grafana.”

“Building and deploying production in-house, domain-specific LLM chatbots for enterprises that cannot use third-party GPT tools due to internal policies. Focused on reducing latency and improving domain awareness using fine-tuning, continual learning, and advanced RAG/agent retrieval strategies, with experience orchestrating multi-agent workflows via LangChain/LlamaIndex and vector DBs (FAISS, Weaviate, Chroma).”

“ML practitioner who built a production customer risk scoring system to replace slow manual approvals, owning the full pipeline from feature engineering and XGBoost training to deploying a Dockerized FastAPI prediction service. Emphasizes reliability and business-aligned evaluation (recall/ROC-AUC, threshold tuning, drift monitoring) and is comfortable translating model decisions into stakeholder metrics like conversion rate (experience at EasyBee AI).”

“Backend/AI engineer who built and refactored a production RAG system over IRS Form 990 filings for 60 nonprofits, using a dual-path architecture (deterministic financial ranking + TF-IDF semantic retrieval) to keep latency sub-2s and reduce hallucinations. Demonstrates strong API craftsmanship in FastAPI (contract-first, OpenAPI-driven) plus production-grade security for multi-tenant systems (JWT, RBAC, Supabase-style RLS) and careful migration practices (feature flags, traffic mirroring, incremental rollout).”

“Robotics engineer who built a real, mostly self-assembled autonomous robot (WRAITH) as a final-year project, implementing ROS2-based 2D SLAM (Cartographer/SLAM Toolbox) and Nav2 on a Raspberry Pi 5 under tight CPU/RAM and OS compatibility constraints. Also delivered a full Flutter mobile control app backed by a Flask REST API (manual control, live camera streaming, mapping/navigation) and introduced an image-based verification method to improve localization.”

“Built and productionized LLM-powered workflows that generate contextual insights from structured financial data, including prompt/retrieval design, data standardization, and reliability controls like rate limiting and batching. Also diagnosed and fixed real-time failures in an automated order validation system using logs/metrics, staging reproduction, edge-case handling, retries, and alerting, while supporting sales/customer teams with demos, scripts, and FAQs to drive adoption.”

“Built and deployed a production RAG-based internal document Q&A system using LangChain, vector search, and a dockerized FastAPI LLM service. Focused on reliability by systematically reducing hallucinations and improving retrieval through prompt grounding/abstention strategies, chunking and top-k tuning, and iterative evaluation with logged metrics and manual validation.”

“ML/NLP engineer who has built production systems end-to-end: a real-time recommendation platform (100k+ profiles) using BERTopic-style clustering and a RAG-based news summarization/recommendation stack with ChromaDB. Strong focus on scaling and reliability (GPU batching, Redis caching, Kafka ingestion, Docker/Kubernetes, Prometheus/Grafana) and on maintaining model quality over time via drift monitoring and retraining triggers.”

“AI/LLM engineer who has shipped production RAG and agentic systems end-to-end (LangChain/FAISS, OpenAI+Gemini, FastAPI, Docker, Streamlit), focusing on retrieval quality and low-latency performance. Also partnered with a non-technical PM at deepNow to deliver a forecasting + summarization pipeline for daily market insights with iterative prototyping and a simple UI.”