ali-asghari / senior-engineer-v6.0
Ali Asghari
Senior Software Engineer & AI Specialist
Building at Careberry
6+ years experience
Python · AI/ML · Backend
Open to Relocation
## Description
About
Senior Software Engineer with 6+ years of experience specializing in
AI-driven applications and high-scale Python backends. Expert in architecting
RAG pipelines, fine-tuning LLM workflows (Qwen, GPT, Gemini),
and deploying production-grade AI services. Proven track record of handling
3M+ daily requests and maintaining high system reliability across
healthcare, fintech, edtech, and food delivery industries.
📍 Open to relocating to Barcelona, Spain
## Capabilities
Technical Skills
🧠 AI / ML
RAG
LLM Guardrails
LLM Orchestration
Ollama
Groq
ChromaDB
Whisper (STT)
Transformers
Prompt Engineering
⚙️ Backend
Python
FastAPI
Django
Flask
Celery
RabbitMQ
RESTful APIs
System Design
🗄️ Data & DevOps
PostgreSQL
Redis
S3
Docker
Docker Swarm
CI/CD
Grafana
Prometheus
📝 Languages
Python
JavaScript (React.js)
SQL
Java
## Fine-tuning History
Work Experience
- Designed a high-reliability RAG orchestration layer for healthcare data, implementing LLM response guardrails and validation checks to mitigate hallucinations and ensure 100% adherence to clinical safety protocols.
- Optimizing LLM inference pipelines to improve response latency and accuracy for internal healthcare workflows.
- Integrating vector-based search to streamline access to large-scale medical knowledge bases.
- Developed an automated Q&A generation pipeline using Whisper and multi-model LLMs, processing video content into structured educational assets.
- Engineered a real-time, voice-enabled Q&A RAG system using ChromaDB and Ollama models, allowing users to query lecture content via natural speech.
- Implemented a Django-based "Chat with PDF" application using Qwen models, enabling context-aware conversations on user-uploaded files.
- Maintained and optimized high-throughput APIs via Flask, handling 3 million daily requests for a user base of 6+ million.
- Redesigned a Django reporting engine, achieving a 66% reduction in generation time for critical financial reports.
- Enhanced system responsiveness by implementing advanced Redis caching strategies and PostgreSQL query tuning.
- Refactored the order-matching engine using the Hungarian Algorithm, resulting in a 6x speed increase for 3,000+ daily deliveries.
- Built an autonomous financial microservice using Celery/RabbitMQ to manage weekly payouts for 1,000+ fleet members.
- Deployed dynamic auto-scaling solutions using Docker Swarm and monitoring dashboards with Grafana and Prometheus, ensuring 99.9% uptime during peak traffic.
## Training Data
Education
MSc in Computer Science
Tarbiat Modares University
Graduated 2022
BSc in Computer Science
University of Tabriz
Graduated 2019