Skip to content
View SrirajBehera's full-sized avatar
🎯
Focusing
🎯
Focusing

Organizations

@LNMIIT-Counselling-Cell

Block or report SrirajBehera

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
SrirajBehera/README.md

Sriraj Behera

Software Engineer · Systems & AI Infrastructure · Bengaluru

FinTech Distributed Systems LLM Infrastructure Full Stack Blockchain


Portfolio LinkedIn Twitter


Software Engineer at Finmo — a global payments and FinTech infrastructure company. Day to day, I design and ship systems that move money reliably: FX hedging engines, cross-border payment connectors, reconciliation pipelines, IAM and auth flows, and multi-organization treasury infrastructure. The domain is unforgiving — correctness, fault tolerance, and auditability aren't nice-to-haves.

My instinct on any problem is architecture-first. Understand the failure modes, define the boundaries, reason about consistency and state, then write the code. I vibe code freely and ship fast, but I hold the underlying design to a high standard. The two aren't in conflict — moving quickly is only sustainable when the system underneath is honest.

Outside of payments, I build in the AI/ML space: RAG pipelines, LLM tooling, and developer-facing systems where retrieval quality and inference performance actually matter.


What I Work On

At work
  ├── FIX protocol trading engines & FX hedging infrastructure
  ├── Idempotent accounting sync, ledger reconciliation & audit systems
  ├── Config-driven payment connector architecture (multi-currency, multi-provider)
  ├── OAuth2.0 / PKCE auth flows, RBAC, AWS Cognito session governance
  ├── Multi-org hierarchy systems — KYB, settlements, fund flows, pricing
  └── Distributed queue processing, event-driven architecture, BullMQ + Redis

Outside
  ├── RAG pipeline infrastructure — indexing, retrieval, cross-encoder reranking
  ├── Streaming speech-to-text engines & real-time audio processing
  ├── Keyboard-driven developer tooling & fast local-first CLI systems
  └── Blockchain / on-chain contract design

What I'm Always Learning

  Distributed systems internals  →  consensus, replication, partition tolerance
  Payments domain depth          →  FIX protocol, settlement rails, FX mechanics
  LLM infrastructure             →  inference optimization, retrieval quality, embedding strategy
  Systems programming            →  memory, concurrency, performance at the metal
  Scale patterns                 →  rate limiting, backpressure, idempotency, exactly-once delivery

Stack

Languages & Runtimes

TypeScript Python JavaScript Node.js C++ Java Solidity

Backend & Architecture

NestJS Express Django BullMQ Microservices Event--Driven

Data & Messaging

MySQL PostgreSQL MongoDB Redis Firebase

AI / ML & LLM Infrastructure

LangChain HuggingFace OpenAI RAG Vector DBs Cross-Encoders

Frontend & Mobile

React Next.js React Native Expo

Cloud & Infra

AWS GCP Docker Kubernetes Jenkins Grafana GitHub Actions

Pinned Loading

  1. fast-explorer fast-explorer Public

    A blazing-fast, keyboard-driven file explorer and search engine for macOS.

    TypeScript

  2. Inumaki-speech-to-text Inumaki-speech-to-text Public

    A high-performance, multi-threaded streaming speech-to-text engine for macOS. This system provides a "Wispr-like" experience: real-time feedback through a frosted-glass HUD, smart semantic endpoint…

    Python

  3. synapseeed/rag-cli synapseeed/rag-cli Public

    A powerful, multi-pass codebase indexing and graph-resolution engine designed to build a high-fidelity Retrieval-Augmented Generation (RAG) system for large, multi-repository microservices architec…

    JavaScript

  4. synapseeed/rag-reranker synapseeed/rag-reranker Public

    A high-performance reranking service designed for Retrieval-Augmented Generation (RAG) pipelines. This service uses Cross-Encoder models to provide precise relevance scores for a set of documents r…

    Python

  5. F1-GPT F1-GPT Public

    An AI-powered Formula 1 assistant that answers anything about F1 — including events beyond the model's knowledge cutoff — using Retrieval-Augmented Generation (RAG), Google Gemini embeddings, and a…

    TypeScript

  6. LNMIIT-Counselling-Cell/mobile-frontend LNMIIT-Counselling-Cell/mobile-frontend Public

    JavaScript