Currently building Relay, a webhook delivery service
I work mostly on backend, data science and low-level programming: REST APIs, backend services, background jobs, data analysis, ML models and systems-level code
Going deeper into machine learning, from the math behind a model to how it learns
Spent a lot of time on low-level programming: memory layout, stack vs heap, pointers and system calls
Studying operating systems and computer architecture to understand what really happens beneath my code
I enjoy solving algorithmic problems, especially dynamic programming, graphs and recursion
I like learning things from first principles, and I keep asking "why?" until I hit the hardware
Ask me about backend, data science and ML, or systems internals: memory management, operating systems and computer architecture
Open to opportunities in backend, data science and ML
A webhook delivery service. It accepts events, delivers them to subscriber endpoints in background workers, and retries failed deliveries instead of dropping them.
flowchart LR
P[Producer] -->|POST event| A[FastAPI]
A -->|store| M[(MongoDB)]
A -->|enqueue delivery| R[(Redis)]
R --> W[Celery worker]
W -->|HTTP POST| S[Subscriber endpoint]
W -. retry on failure .-> R
classDef ext fill:#ff9e64,stroke:#ff9e64,color:#1a1b27
classDef api fill:#7aa2f7,stroke:#7aa2f7,color:#1a1b27
classDef store fill:#9ece6a,stroke:#9ece6a,color:#1a1b27
classDef queue fill:#bb9af7,stroke:#bb9af7,color:#1a1b27
classDef worker fill:#7dcfff,stroke:#7dcfff,color:#1a1b27
class P,S ext
class A api
class M store
class R queue
class W worker
To me, software is a stack of layers. I build at the top, whether that's a backend, an API or an ML model. Then I follow my code down one layer at a time, asking what runs it, who manages it and what finally executes it, until I understand what the machine is actually doing.
⬆️ building software · understanding computers ⬇️
Languages
Backend & databases
Data science & ML
Tools