Looking for a job

Sofiya Chobanyan

@sofiyachobanyan

Lead Backend & AI Engineer: production LLM, RAG & AI agents on Python/AWS. Cut AI inference cost 86% in production. Claude Certified Architect.

Armenia

Message

What I'm looking for

I’m looking for a remote Lead/Senior role building production AI systems: LLM pipelines, RAG, and AI agents on a solid Python backend. I care about measurable impact (my favorite metric: cutting AI inference cost 86% while holding quality), owning work from experiment to production, and product teams over client-facing consulting.

I’m a Lead Backend & AI Engineer with 5+ years shipping production Python systems across healthcare, fintech, and legal tech, moving from backend-only REST services to full AI + backend integrations.

I build AI systems that survive production. On a healthcare data platform serving 50,000+ patients a month, I architected the document intelligence pipeline end to end: FHIR ingestion from major EMR systems (Epic, Athena, Allscripts, ModMed, and more), OCR, embedding and chunking strategy, similarity search tuning, and batched LLM extraction agents that turn unstructured clinical documents into hundreds of structured parameters per patient. I built two production AI agents: a Claude Agent SDK agent that translates natural language into safe, validated SQL for cohort filtering, and a patient-chat agent combining structured tools with patient-scoped RAG.

The numbers I’m proudest of: cut AI inference cost per patient by 86% ($0.70 to $0.10) through model routing and prompt design, and took key PostgreSQL queries from 3,000ms to 10-15ms. Earlier I shipped LLM systems in fintech (bank statement parsing with anti-hallucination guards) and legal tech (contract extraction with word-level citations), led a cross-functional team of 9, and maintained ~95% test coverage with full CI/CD.

MS in Mathematical Modeling. Claude Certified Architect (Anthropic). Remote-first and async since 2020. Looking for Lead/Senior Backend & AI roles on product teams.

Experience

Work history, roles, and key accomplishments

Current

Lead Backend & AI Engineer

Current

ScrumLaunch

Feb 2022 - Present (4 years 7 months)

Delivered production backend and AI systems across healthcare, fintech, and legal clients using Python, FastAPI, and AWS. Led a 9-person cross-functional team. Built a FHIR healthcare platform serving 50,000+ patients/month, LLM/RAG document parsing, two production AI agents, and legal contract extraction. Cut AI inference cost 86% and key query latency from 3,000ms to 10-15ms.

Django Developer

Talent500

Oct 2021 - Feb 2022 (4 months)

Built RESTful backend workflows with Django and Django REST Framework and delivered security-compliant APIs for multiple high-traffic systems. Streamlined release times with end-to-end CI/CD automation using Docker and GitHub Actions.

OAuth Django Django REST Framework Python REST APIs Docker GitHub Actions CI CD Security Compliance

Backend Developer (Flask, Django)

Freelance

May 2021 - Oct 2021 (5 months)

Designed, coded, and deployed functional backend solutions for clients using Flask and Django. Improved client satisfaction through documented technical decision-making and supported agile service architecture adaptation.

Flask Django Python Authentication Data Validation Agile Development Technical Documentation

Web Scraping Engineer

Freelance

Sep 2020 - May 2021 (8 months)

Engineered automated scraping workflows to retrieve structured and unstructured data from multiple platforms and ticketing systems. Scaled scraping throughput using asynchronous processing and improved automation efficiency with modular Python scripts.

Python Selenium BeautifulSoup AsyncIO AIOHTTP Data Extraction Web Scraping