I build dependable backend systems and infrastructure tooling with an emphasis on failure modes, testability, and useful evidence.
- Distributed backend services, event-driven systems, and data-intensive workflows
- Reliability engineering, observability, and safer infrastructure changes
- Protocol interoperability and developer-facing tooling
- Production open-source contributions in Python and Go
A lightweight Go-based WebRTC Selective Forwarding Unit built on Pion, with WebSocket signaling, RTP/UDP media forwarding, Prometheus metrics, and Docker/Kubernetes-oriented deployment.
Go · Pion WebRTC · WebSockets · Prometheus · Docker
An open-source FinOps platform for cloud-cost ingestion, forecasting, anomaly detection, background workers, and natural-language analysis.
Python · FastAPI · PostgreSQL · Redis · RabbitMQ · Next.js · OpenTelemetry
An infrastructure pre-deploy gate that parses Terraform and Kubernetes changes, evaluates deterministic guardrails, computes blast-radius and deployment risk, and generates corrected patches through an OpenAI-compatible model and MCP server.
Python · FastAPI · SQLAlchemy · TypeScript · Next.js · Terraform · Kubernetes · MCP
- Apache Airflow #70794 — Fix Cloud Build operator template preparation. Open
- Supabase #42441 — Prevent authentication forms from overwriting a stale
SITE_URL. Open
| Area | Technologies |
|---|---|
| Languages | Go, Python, Java, C++, JavaScript, TypeScript |
| Backend | FastAPI, Node.js, Express, REST APIs, microservices, WebSockets |
| Data & messaging | PostgreSQL, MySQL, MongoDB, Redis, Kafka, RabbitMQ |
| Cloud & infrastructure | AWS, Docker, Kubernetes, Terraform, Nginx |
| Observability | Prometheus, Grafana, OpenTelemetry, CI/CD, automated testing |
| AI/ML | OpenAI and Gemini APIs, TensorFlow.js, forecasting, anomaly detection |
- CodeChef: 4★, peak rating 1951
- I use algorithmic problem-solving to strengthen the fundamentals behind systems work.
- LinkedIn: linkedin.com/in/abhisek343
- GitHub: github.com/abhisek343