Saqib S.
GenAI Engineer | RAG & LLM Systems in Production | Python, FastAPI
RAG pipelines, multi-agent systems, and production LLM applications — built end to end on FastAPI, LangGraph, and Pinecone. 15+ years shipping SaaS platforms from architecture through deployment, US-based. WHAT I'VE SHIPPED Claude-powered executive assistant agent — autonomous task and schedule management. 7 tools wired to calendar and email, 25% task-completion rate on a held-out eval set, 10s median response. Running 500+ actions/day in production across a fleet of users. Multi-agent SaaS platform — connecting unscrupulous financial data to intelligent decisions made in real-time . 15 specialized agents coordinated through LangGraph, 55% reduction from original manual processing pipeline, integrated across 4 third-party platforms. WHAT I BUILD Retrieval systems — grounded answers over PDFs, SQL, and internal docs, with citation and refusal behavior validated against a held-out eval set rather than assumed. Agentic workflows — LangGraph and CrewAI agents that run multi-step tasks with retry, state persistence, and human-in-the-loop checkpoints wherever being wrong is expensive. Full-stack SaaS — auth, Postgres/Supabase, Stripe, and the AI layer wired together and deployed. MVP through production, one engineer end to end. HOW I WORK Every AI system I ship comes with an eval harness. If we can't measure retrieval quality or task success, neither of us can tell whether the last change helped. I write the eval set before the pipeline. I own the whole stack, so integration failures don't turn into a scheduling problem across three contractors. Architecture, security, deployment, and the AI logic sit with one person. Based in Texas, so there's real overlap with US business hours. STACK AI & LLMs: Claude · GPT · Gemini · Llama · Mistral Orchestration: LangChain · LlamaIndex · LangGraph · CrewAI Retrieval: Pinecone · Chroma · FAISS · Weaviate Backend: FastAPI · Django · Flask · Node.js Frontend: React · Next.js · TypeScript · Tailwind Data: PostgreSQL · MongoDB · Redis · Supabase · Firebase Infra: AWS · Docker · CI/CD Tell me what you're building and what "working" looks like for it, and I'll tell you how I'd architect it before you hire anyone.
Earnings and projects
Last 6 months ÷ 6
Last 6 months: $494
Billed in all: $1,564$1.6k
$494 in the last 6 months
Across 3 contracts
Across 10 contracts
Earnings, last 30 days
Skill rankings: Top 26% in MongoDB
Ranked by MRR among everyone on Upwork with the skill.
- #808Natural Language Processing2,688 freelancers · 🇺🇸 #79Top 30%🇺🇸#79 of 264
- #1,577RESTful API6,064 freelancers · 🇺🇸 #109Top 26%🇺🇸#109 of 324
- #1,898FastAPI4,337 freelancers · 🇺🇸 #154Top 44%🇺🇸#154 of 267
- #2,512Docker8,761 freelancers · 🇺🇸 #149Top 29%🇺🇸#149 of 525
- #2,803MongoDB10,865 freelancers · 🇺🇸 #127Top 26%🇺🇸#127 of 522
- #2,883Machine Learning8,972 freelancers · 🇺🇸 #374Top 32%🇺🇸#374 of 1,150
- #4,065Full-Stack Development9,452 freelancers · 🇺🇸 #369Top 43%🇺🇸#369 of 767
- #4,455PostgreSQL12,850 freelancers · 🇺🇸 #323Top 35%🇺🇸#323 of 807