GenAI Engineer · Available · Open to relocate

I build production GenAI that quietly runs the business.

GenAI Engineer with 4+ years building production LLM systems — chatbots, RAG pipelines, email agents, and observability. Delivered insurance chatbots handling ~30K monthly customer interactions at 70% resolution, multi-market email automation at ~99% auto-processing, RAG at 98% precision, and a GenAI Studio that cut chatbot R&D from weeks to days.

Portrait of Ramsis Hammadi

— Ramsis Hammadi, Tunisia · Remote

📍 Tunisia · Remote · Open to relocategithub/Rams901linkedin/ramsis-hammadiramsishammadi@gmail.com

By the numbers

70%

Chatbot resolution rate

~30K

Customer interactions / month

~99%

Email auto-processing

98%

RAG context precision

3×

Faster feature deployment

5.0

Upwork client rating

Flagship — Ominimo

Insurance GenAI that customers actually use.

Problem
Policyholders across Hungary, the Netherlands, and Poland needed self-serve answers — while ops was buried in inbound email.
What I built
Multi-turn insurance chatbots with intent routing and escalation, plus LLM email agents that classify, extract, and file into local insurance IT. A FastAPI studio so the next market does not take weeks.

70%

End-to-end resolution

~30K

Interactions / month

~99%

Email auto-processed

86%

Positive feedback

Full Ominimo role →
Bo insurance chatbot: login, policy actions, and a message field in Dutch
Bo — production insurance assistant, Netherlands

Product and observability

Bo chatbot offering to auto-resolve a policy issue before escalating to a human agent
Human in the loop — auto-handle first, then escalate
Bo chatbot showing unpaid invoices with pay-now actions
Structured actions — invoices, not just chat bubbles
Feedback dashboard with average rating, promoters, detractors, and handoff share
Feedback KPIs — rating, promoters, handoff share
Analytics charts for discussions per day, resolutions, and returning customers
Ops view — volume, resolution, returning customers

Also on the résumé

More engagements.

Full résumé →

Aug 2024 — Nov 2024

Remote · US

Freelance AI Specialist — DCF Copilot

  • —Built a GenAI copilot for real-time DCF models, integrating LLM reasoning with structured financial calculations.
  • —Designed a RAG pipeline with Guardrails (jailbreak prevention) and integrated Weave for LLM evaluation, reaching 98% context precision.

Jan 2024 — Jun 2024

Remote · US

Data Research Science Intern (MSc. End of Studies Project) — Seeloz Inc.

  • —Trained deep RL models using DQN and PPO with StableBaselines3 and Ray RLLib for supply chain optimization.
  • —Integrated and tested 4 novel features for the AI Inference Engine, improving stability and prediction accuracy.

Projects

Also shipped independently.

All work →

Jun 2024 — Aug 2024

Micro-Testing Environment for Agentic Synthesis

Dediro

Freelance consulting

Built a sandboxed testing harness for agentic synthesis pipelines (LangGraph, CrewAI, AutoGen), enabling rapid iteration on agent orchestration and prompt chains — accelerated feature deployment by 3×.

Dec 2023 — Jun 2024

Survey Data Strategy

Unilever, Inc.

Freelance engagement

Ingested and analyzed 10K+ survey responses (PCA, MDR, EDA). Delivered interactive dashboards and recommendations projecting 25% satisfaction increase.

Oct 2023 — Mar 2024

AI-Powered SEO Article Generator

GM, Inc.

Freelance project

Built an LLM-powered tool generating SEO-optimized articles with auto-generated images, integrated with WordPress for one-click publishing and internal linking.

Word of mouth

What collaborators say.

“Ramsis is incredible and the best thing to happen for my company. I cannot recommend him enough. He is extremely smart, driven, and cares deeply about his work.”
Farzad Khosravi · Founder, Dediro
“Ramsis is likely going to change the world with data and machine-learning. He went quickly from chunk-based to chained to a sophisticated multi-agent model which blew us all away.”
Christian Lilley · CTO, Ex-McKinsey
“Ramsis consistently impressed our team with his profound understanding of AI and his capacity to devise inventive solutions for intricate problems.”
Dario Gomez · Senior Software Engineer
“Ramsis created an efficient app that did exactly what I wanted. He's very good at tying AI processes together — always quick to answer, pivot, and accommodate ideas.”
Karianne Gagnon · Founder & CEO, GM, Inc.

Let's build

Have a GenAI problem worth solving?

Open to full-time roles (remote or relocate) and select freelance engagements in GenAI, RAG, and agentic systems.