About

Jaber Ahmed Deedar — BUET Computer Science & Engineering (B.Sc., ongoing; CGPA 3.91, Dec 2021–present). Previously Notre Dame College (HSC, Jul 2018–Apr 2020).

Engineer working across full-stack product delivery, backend and distributed systems, and applied AI (RAG, vector search, agentic workflows). Practice includes cybersecurity through CTF and web security work alongside shipping production features.

Primary focus: productionizing AI, robust APIs and async pipelines, and shipping under startup-style constraints—from ed-tech at scale to AI platform backends.

What I optimize for

• Reliable async orchestration — Clear boundaries between APIs, queues, and workers; Redis/Kafka-style patterns; workflow resume and tool-call loops.
• Evidence-backed AI — Domain-specific training and evaluation (large QA sets, retrieval quality), not demo-only LLM wrappers.
• Operations under fire — DoS and abuse, DB and disk incidents, live traffic spikes (live classes, viral launches).
• Security-aware engineering — OIDC/JWT/API-key patterns in product backends; CTF-driven offensive/defensive intuition.

Public profiles

• GitHub: selen0phile
• Kaggle: selen0phile — Contributor tier; 132 public kernels
• Codeforces: selenophile — max rating 1714; 2000+ solved problems across online judges