AI/ML Engineer who ships production AI products end-to-end.

I build LLM + data systems that turn messy business workflows into reliable automation: retrieval, evals, deployment, and monitoring.

Built 3 SaaS products (live demos below) LLM apps: RAG, agents, prompt/eval pipelines Strong analytics foundation (SQL/Python/Tableau)
Download Resume See Case Studies ↓
Soya Diaoune
Quick Fit Check
  • Target roles: AI/ML Engineer, Applied ML, LLM Engineer
  • Domains: B2B SaaS automation, marketplaces, HR tooling
  • Location: Charlotte, NC — Open to remote
ATS Keywords
Python TypeScript SQL React Node.js LLMs RAG OpenAI Supabase PostgreSQL AWS Tableau

Featured Case Studies

SponsorSynq

Event sponsorship platform with AI-powered matchmaking

Live
Problem
  • Manual sponsor-event matching took weeks
  • Proposal creation was slow and inconsistent
What I Built
  • Matching engine for sponsors ↔ events
  • LLM-powered proposal generator
  • Contract automation workflow
  • Integrated payment system
ML/AI Details
  • Ranking algorithms for match quality
  • Lead scoring with historical data
  • LLM proposal generation with guardrails
  • ROI reporting and analytics
Tech Stack
  • Next.js, Supabase, OpenAI
  • Vector embeddings for matching
  • Stripe for payments
End-to-end automation, faster deal cycles, reduced manual work

Staffless

AI Core for operators across business functions

Live
Problem
  • Fragmented workflows across teams
  • Manual operations consuming resources
What I Built
  • Multi-agent orchestration system
  • Tool-calling layer for integrations
  • Call and chat automation
  • Admin dashboard for monitoring
ML/AI Details
  • Multi-agent workflows with routing
  • Tool orchestration and fallbacks
  • Quality + safety evaluations
  • Real-time response streaming
Tech Stack
  • Next.js, FastAPI, OpenAI
  • WebSockets for real-time
  • Custom evaluation pipeline
Unified AI operations layer, improved team efficiency

autoHR

AI HR chatbot trained on company policies

Live
Problem
  • Repetitive HR questions overwhelming teams
  • Employees couldn't find policy answers
What I Built
  • RAG pipeline over HR documents
  • Citation system for source tracking
  • Admin analytics dashboard
  • Multi-tenant architecture
ML/AI Details
  • RAG over HR policies + handbooks
  • Citation-first answer generation
  • Hallucination controls + guardrails
  • Query analytics for HR insights
Tech Stack
  • Next.js, Supabase pgvector
  • OpenAI embeddings + chat
  • Document parsing pipeline
Instant policy answers with sources, reduced HR ticket volume

How I Build ML Systems

RAG / Knowledge QA

Ingestion → Chunking → Embeddings → Retrieval → Answer + Citations → Eval Set

LLM Reliability

Prompt Versioning → Test Prompts → Offline Evals → Red-team Checks → Fallbacks

Deployment

API (FastAPI) → Caching → Queues → Monitoring → Cost Tracking

Selected Open Source

spotifAI
Prompt → structured music generation pipeline
OpenAI • music21 • MIDI output

Experience

2023 – Present
Founder & AI/ML Engineer
Independent (SponsorSynq, Staffless, autoHR)
  • Built and shipped 3 production AI/ML products from zero to live users
  • Designed RAG pipelines, multi-agent systems, and LLM evaluation frameworks
  • Implemented end-to-end ML infrastructure: embeddings, vector search, prompt engineering
Python Next.js OpenAI Supabase FastAPI pgvector
2025 – Present
AVP; Senior Finance Analyst
Bank of America
  • Mapped full forecasting lifecycle, identified data-quality gaps, implemented automation and control enhancements
  • Reduced manual effort and process risk while strengthening auditability and model lifecycle controls
  • Previously: QA Consultant V — executed enterprise-wide model testing for ML-enabled fraud, AML, and compliance controls
SQL Python AI/ML Governance Risk Analytics
2023 – 2025
Risk Analyst
Coca-Cola Consolidated
  • Developed dashboards using Tableau and Power BI; managed extensive risk data with SQL, Python, Excel
  • Automated Loss Summary Data reporting with Python and OfficeScript, reducing report generation time
  • Designed SQL scripts for data validation, anomaly detection, and trend analysis in risk reporting
SQL Python Tableau Power BI Ventiv
2019 – 2022
Systems Operations Senior Analyst
Wells Fargo
  • Designed and implemented automation scripts (bash shell scripting), replaced manual processes
  • Fixed Python & Java (Spring) back-end bugs; improved frontend React UI
  • Analyzed systems operations processes, troubleshot complex UNIX/Linux problems
Python Java React Bash UNIX/Linux
2017 – 2019
B.S. Computer Science
Florida International University, Miami, FL
  • 3rd Place at HackStetson (Study Buddies)

See full history on LinkedIn →

Want a candidate who ships?

Download Resume Email Me LinkedIn
soyadiaoune@gmail.com | linkedin.com/in/soya-diaoune | github.com/sdiaoune