Vetted OpenAI API Professionals

“Hands-on Engineering Manager at Hyper-Reach modernizing a mission-critical emergency alerts platform used by county emergency directors and first responders during live disasters. He led a phased migration from a 25-year-old JSP monolith to React, shipped in 5 months instead of an estimated 1.5 years, with zero regression incidents and clear customer and sales impact.”

Fadi Abbas

Screened ReferencesStrong rec.

Executive product leader specializing in AI-native products and EdTech platforms

Dearborn Heights, MI16y exp

“Product leader and founder with experience spanning EdTech, AI, LegalTech, and mobile imaging. He co-founded a tutoring platform that grew to 300K+ students and later built citation-grounded legal AI systems designed for high-trust, high-stakes workflows. Particularly compelling is his consistent philosophy of using AI to amplify human expertise rather than replace it, backed by global team leadership and strong product execution across EMEA, the US, and Asia.”

“Built and deployed a production LLM-based RAG agent platform adopted by multiple business teams (Marketing, GTM, Recruiting, Customer Support) to automate knowledge search, Q&A, and content generation. Emphasizes production-grade reliability (grounding/validation/guardrails), rigorous evaluation/monitoring, and cost-aware scaling via model tiering, prompt/retrieval optimization, and caching using LangChain/LangGraph orchestration.”

“Built an LLM-powered academic research assistant for a professor (LangChain + OpenAI + arXiv) focused on synthesizing papers quickly, with emphasis on reliability (ReAct prompting, citation verification) and cost control (caching). Has production MLOps/orchestration experience at Cisco and HCL Tech using Kubernetes, plus MLflow and GitHub Actions for lifecycle management and CI/CD.”

“Built a secure, on-prem/private GPT assistant to replace manual SharePoint-style search across thousands of policies/SOPs/engineering docs, using a production RAG stack (LangChain/LangGraph, FAISS/Chroma, PyMuPDF+OCR, vLLM). Implemented layout-aware ingestion (including table-to-JSON) and a multi-agent retrieval/generation/verification workflow with strong observability and compliance guardrails, delivering ~70% reduction in search time.”

“Serial entrepreneur since leaving corporate in 2009, working largely for equity on multiple startups. Building (1) academically rigorous, anti-overfitting quant/backtesting tools for retail investors (with potential applicability to smaller hedge funds lacking quant staff) and (2) a partner-led “social-as-a-service” platform for verticals like real estate/PropTech (including FSBO use cases) focused on first-party data capture vs. big tech.”

“Full-stack engineer who built a cloud storage app feature (file upload/management) with Next.js App Router + TypeScript and owned post-launch improvements. Also has internship experience building a geospatial AI chatbot: designed Postgres/PostGIS data models and optimized spatial queries, and implemented an LLM workflow orchestrated with LangChain/LangGraph plus a RAG pipeline grounded in OpenStreetMap data to reduce hallucinations.”

“Data engineer focused on AWS-based enterprise data platforms, owning end-to-end pipelines from multi-source batch/stream ingestion (Glue/Kinesis/StreamSets/Airflow) through PySpark transformations into curated datasets for Redshift/Snowflake. Emphasizes production reliability with strong monitoring/observability and data quality gates, and reports ~30% performance improvement plus improved SLAs and latency after optimization.”

Chandan Chalumuri

Screened

Mid-level Data Scientist specializing in ML, NLP, and Generative AI

Tempe, AZ4y exp

“Data engineering / ML practitioner with experience at MetLife building transformer-based sentiment analysis over large unstructured datasets and productionizing pipelines with Airflow/PySpark/Hadoop (reported 52% efficiency gain). Also implemented embedding-based semantic search using Pinecone/Weaviate to improve retrieval relevance and enable RAG for customer support and document matching use cases.”

RAJ KUMAR

Screened

Mid-level Full-Stack Software Engineer specializing in FinTech and cloud-native microservices

Chicago, IL6y exp

“Backend engineer at Discover who built and scaled Python/Flask services for a card dispute resolution platform, tackling long-running external network validations with Celery+Redis and delivering measurable gains (response time ~3s to <300ms; throughput +40%). Experienced in high-scale PostgreSQL/SQLAlchemy optimization (partitioning, read replicas, N+1 avoidance), event-driven systems with Kafka, and integrating ML fraud detection using AWS SageMaker/Lambda/ECS with clear separation of real-time vs batch processing.”

Manpreet Kour

Screened

Senior Data Scientist specializing in Generative AI and NLP

Seattle, USA6y exp

“ML/NLP engineer with recent Scotiabank experience building production-grade indexing automation over large-scale emails and customer databases, combining LLM fine-tuning (Mistral, XLM-R) with fuzzy matching to exceed 95% accuracy under strict banking constraints. Also built a RAG-based chat agent using Gecko embeddings, Vertex AI Search, Gemini, and cross-encoder reranking, and delivered a text-to-SQL chatbot at SOTI through iterative fine-tuning and benchmark-driven experimentation.”

Madhu Sriram Sengottu Velan

Screened

Entry-Level Software Engineer specializing in backend systems and distributed services

Chennai, IN2y exp

“Backend/AI engineer from an early-stage Japan-based startup (WorkAI) who built a multi-tenant RAG system integrating Notion/Slack/Google Drive with Pinecone and OpenAI, including a chatbot retrieval workflow. Experienced in production reliability (rate limits, retries, verification layers), strong Python/FastAPI engineering practices, and PostgreSQL performance optimization; currently based in India and needs sponsorship.”

“Built a Python pipeline to monitor and classify public posts from sources like Hacker News and Reddit for SWE/tech job opportunities, with a strong focus on reliability, observability, and recoverable failures. Also currently building a court queueing system for the UCSD Badminton Club, showing an ability to turn messy, informal real-world processes into practical automation through iterative user feedback.”