Sina AmarehavailableAsk me anything about my work.
From multi-provider RAG systems to crash-tolerant backends — ask me about anything I've built.
Things I've built
Production-minded open-source projects where backend resilience meets LLM engineering.
Turns your own Telegram account into a self-hosted, installable AI messenger (formerly SakaiBot). Read and send messages with inline media in a glassy web app, and call an LLM right inside any chat — analyze, ask, translate, image, voice. AI results land in a panel, so you decide what to send.
A browser extension that turns a rough draft into an engineered prompt inside any text field on any site — with your own API key. The result appears next to your original, nothing is replaced until you accept, and pasted code comes back byte-for-byte. Live on Firefox Add-ons.
A bit about me
I'm a Python developer with a Computer Science degree, focused on backend services and LLM-powered applications. Alongside that I work with Mercor as an AI training & evaluation contractor — designing adversarial prompts and tasks that find exactly where frontier models break. I learn mainly by building — my open-source projects are where ideas get tested and pushed toward production quality. I'm early in my career and still growing, but steady about shipping, writing tests, and seeing work through.
- Applied AI/LLM work — RAG, LangGraph, and multi-provider failover behind FastAPI
- Frontier-model evaluation — adversarial prompts, golden answers, and rubrics that make failures reproducible
- Resilient backends — jobs are leased and watchdog-recovered, so a crash means a resumed run, not lost work
- Quick from prototype to production, with tests and clean architecture
“Never had to explain something twice. Never came back empty-handed. Your biggest strength? Being a problem solver.”
AI Training & Evaluation (Contract) · Mercor
2026 · presentAdversarial evaluation of frontier LLMs — designing prompts and tasks built to make strong models fail on web search, multi-step reasoning, and the edge cases where they're confidently wrong, then writing verified golden answers and single-answer rubrics so every failure is reproducible and objectively scorable.
Software Developer · Dekamond
2025 · 6 monthsBuilt AI & automation for Kaleri.ai — wrote multiple automation systems and data-scraping pipelines that collect and process data, plus multi-provider LLM workflows (LangGraph, RAG) behind FastAPI for BI tooling and internal developer tools, and cut LLM running costs ~70% with model routing, caching, and prompt optimization.
Django Developer (Backend) · Arnikup
2024 · 6 monthsBuilt and maintained REST APIs for a food-delivery platform in a five-person team — modeling catalog and orders in PostgreSQL with Django REST Framework, offloading background work like order processing and notifications to Celery, using Redis for caching and as the task broker, containerizing services with Docker, and testing with pytest.
M.Sc., Software Engineering
2025 – presentIslamic Azad University — Science & Research, Tehran
B.Sc., Computer Science
2020 – 2024University of Guilan, Rasht
- 2nd place — team Python programming competition, University of Guilan (2023)
- Undergraduate research on Particle Swarm Optimization
- English (professional working proficiency) · Persian (native)
Let's build something together
I'm open to backend & AI engineering roles and interesting collaborations. Send a note — it reaches me instantly — or email me directly.