I'm Prathmesh. I build — double-entry ledgers with strict multi-tenant isolation, a 20-plus-chain perpetuals exchange with 50x leverage and sub-second order streams, and indexers that have processed close to a billion on-chain events, — LangGraph research harnesses and RAG pipelines evaluated against live market feeds and retrieval benchmarks rather than static tests, plus an autonomous Telegram trading assistant reading RSI, MACD, and order flow, and — ingestion pipelines sustaining 120 million events a day across six schemas, sub-second Redis/Socket.io fan-out, and TimescaleDB architectures tuned to reclaim 2.4 terabytes across 379 compressed chunks.
Same information, three shapes. A recruiter skims the page, a terminal person runs curl, an agent reads the structured file.
Backend, Blockchain, and Agentic AI Engineer with 2+ years of experience building high-throughput data platforms, real-time exchange infrastructure, multi-tenant financial ledgers, and autonomous LLM systems. I build systems where correctness and scale matter: ingestion pipelines sustaining 120M events/day (nearly 1B total events), 2.4 TiB of hypertable storage reclaimed, double-entry ledgers, 20+ chain settlements, and four ETHGlobal prize tracks won. Currently pursuing an M.S. in Financial Engineering at WorldQuant University.
Engineering lead for Ledgends, a LINE-native household financial-learning platform featuring multi-tenant architecture, a double-entry ledger, OCR receipt intelligence, and replaceable EVM anchoring.
Engineered the market-data and indexing infrastructure for Assymetrix, aggregating Polymarket, Kalshi, and Limitless into a unified prediction-market data platform sustaining 120M events/day.
synchronous_commit=off, chunked bulk upserts, targeted indexes, tuned Subsquid batch parameters, single-pass aggregate refreshes); rewrote bottleneck endpoints and hardened backfill tooling.Architected GVEX, a hybrid decentralized perpetual exchange aggregating native and Hyperliquid liquidity with up to 50x leverage, multi-chain settlement across 20+ chains, and an AI trading bot.
Integrated 40+ DeFi protocols across 12+ EVM networks into a solver framework, built Tenderly simulation pipelines, hourly AWS Lambda ETL streams, and mentored 9 interns.
Conducted deep quantitative and qualitative research on Solana restaking, Bitcoin scaling economics, and market structure, publishing decision-useful institutional reports.
Researched smart contract security vulnerabilities and proxy upgrade patterns, conducted on-chain incident forensics, and authored the $1.6M Pike Finance exploit postmortem.
delegatecall mechanics, EIP-1967, and Transparent, UUPS, Beacon, Diamond, and minimal proxies, detailing testing approaches for storage collisions and uninitialized proxies.Master's Coursework Financial Markets, Financial Econometrics, Derivative Pricing, Stochastic Modelling, Machine Learning, Deep Learning, Portfolio Management, Risk Management.
Undergraduate Coursework Data Structures and Algorithms, Computer Networks, Blockchain & Smart Contract Security, Mathematics, Electronics, Communication Principles.
TypeScript, Python, Solidity, JavaScript, SQL, Java, Bash, Move
Node.js, NestJS, Fastify, Hono, FastAPI, Uvicorn, REST API Design, WebSockets, Server-Sent Events (SSE), Socket.io, Webhooks (HMAC SHA-256), JWT, Pydantic, Microservices Architecture, Containerized Systems
PostgreSQL, TimescaleDB / TigerData, Redis, DynamoDB, MongoDB, SQLite, pgvector, Qdrant, Pinecone, ChromaDB, PostgreSQL RLS, Kysely, TypeORM, SQLAlchemy, asyncpg, Psycopg, ETL Pipelines, Data Normalization
LangGraph, LangChain, Multi-Agent Systems, Agentic AI, Model Context Protocol (MCP), Tool Calling, Retrieval-Augmented Generation (RAG), Cross-Encoder Reranking, Human-in-the-Loop (HITL), LLM Evaluation & Benchmarking, Gemini, Groq, Mistral, OpenAI Whisper, Sarvam AI
Ethereum / EVM, Polygon, Sui, Smart Contracts, Foundry, ethers.js, web3.py, ERC-4337 (Account Abstraction), ERC-1155, EIP-712, UUPS, Chainlink CCIP, Subsquid (SQD), Goldsky, Chainstack RPC, LI.FI, Safe, Tenderly, Reorg-Aware Indexing
Double-Entry Ledgers, Exchange Infrastructure, Trading APIs, Market Data & Order Books, Perpetuals (up to 50x leverage), Prediction Markets, Paper Trading, Risk Controls, Cross-Border Payment Rails, Treasury Operations, Financial Econometrics, Derivative Pricing, Stochastic Modelling
AWS (Lambda, API Gateway, RDS, DynamoDB, CloudWatch, EC2, ECR), DigitalOcean, Alibaba Cloud, Docker, Testcontainers, GitHub Actions, CI/CD, PM2, Grafana, Prometheus, OpenTelemetry, pnpm, Turborepo
Vitest, Playwright, pytest, On-Chain Simulation (Tenderly), Invariant & Property-Oriented Testing, Load Testing, Unit/Integration/E2E, AES-256-GCM, HMAC Validation, Subprocess Sandboxing, Circuit Breakers
# Prathmesh Ranjan > Backend, Blockchain and Agentic AI Engineer > M.S. Financial Engineering (WorldQuant University, Expected Oct 2027) > B.E. Electronics & Communication (LNMIIT Jaipur, Graduated May 2025) > Contact: prathmeshranjan@gmail.com > Profiles: github.com/PrathmeshRanjan · linkedin.com/in/prathmesh-ranjan ## Professional Experience - Sodality (Jul 2026 - Present | Remote) Lead Backend & AI Engineer / AI & Blockchain Engineer • Ledgends: LINE-native household financial-learning platform. • Multi-tenant backend (TypeScript, NestJS/Fastify, PostgreSQL) with tenant isolation, household identity, guardian permissions, and consent controls. • Double-entry ledger handling exact THB and virtual balances, OCR receipt processing, payment evidence, and replaceable EVM anchoring. • Modular, cloud-portable architecture deployable on DigitalOcean and Alibaba Cloud. - Hashmark Labs (Mar 2026 - Jul 2026 | Remote) Backend & Data Infrastructure Engineer / Blockchain & Backend Engineer • Assymetrix: Multi-venue prediction-market platform aggregating Polymarket, Kalshi, and Limitless. • Dual Polygon indexers (Subsquid / SQD and Goldsky) in TypeScript/NestJS, indexing nearly 1 billion events into PostgreSQL via canonical write paths (TypeORM, Chainstack RPC). • Canonical Data Normalization Layer: 2-level PostgreSQL schema unifying markets, trades, order books, OHLCV candles, and freshness signals into a single API. • PostgreSQL/TimescaleDB ingestion tuning sustaining ~120M events/day across 6 schemas (>3x throughput gain via synchronous_commit=off, bulk upserts, targeted indexes). • TimescaleDB storage strategy: reclaimed 2.4 TiB (4.3 TiB down to 1.9 TiB, 56% cut) with zero failures across 379 compressed chunks. • Sub-second Redis pub/sub and Socket.io fan-out; PM2, Grafana, and Prometheus monitoring. - GPUNET (Nov 2025 - Mar 2026 | Bengaluru) Backend Engineer & Product Lead / Blockchain Backend Engineer • GVEX (gpu.net/gvex): Hybrid perpetual DEX aggregating native and Hyperliquid liquidity into one trading interface. • TypeScript, Hono, PostgreSQL, Redis backend with JWT wallet auth, real-time SSE price/order streams, market/limit/TP-SL/reduce-only orders, cross/isolated margin, and perpetual leverage up to 50x. • Hyperliquid HIP-3 integration (equities, commodities, FX) and multi-chain deposit/withdrawal flows across 20+ chains via LI.FI, Alchemy webhooks, and treasury automation. • Telegram AI trading agent using Gemini and Groq for natural-language trade parsing, RSI, MACD, Bollinger Bands, conditional-order monitoring, and AES-256-GCM encryption. • Integrated native chain and GVEX with DeFiLlama for public TVL and ecosystem reporting. - GlueX Protocol (Aug 2024 - Jul 2025 | Remote) DeFi Software Engineer / Software Engineer • Integrated 40+ DeFi protocols (Compound v3, DODO v2, Morpho) across 12+ chains into solver framework; replicated AMM logic in Python, calculated lending/borrowing APY. • Tenderly-backed end-to-end testing framework verifying off-chain financial math against on-chain simulations for correctness and data integrity. • Hourly AWS Lambda ETL pipelines indexing blockchain events across 12+ chains; managed production RDS and DynamoDB datasets (SQL, Pandas, SQLAlchemy). • Mentored team of 9 interns through onboarding and production protocol integrations. - PYOR (Jul 2024 - Aug 2024 | Remote) Protocol Researcher and Writer • Qualitative & quantitative research on Bitcoin, Solana, DeFi, restaking, L2 networks, tokenomics, and market structure. • Solayer Restaking: Traced deposits from $23.5M to $100M+ in <2 months; analyzed epoch growth, LST composition, operator delegation, slashing exposure, and stake-weighted QoS vs EigenLayer. • Bitcoin Scalability Series: 2-part research series assessing blockspace congestion, fees, miner economics, Lightning liquidity, BitVM, Stacks, and L2 rollups. - QuillAudits (Apr 2024 - Jul 2024 | Remote) Web3 Security Researcher and Writer • Smart contract security research repositories with practical Solidity examples (55+ GitHub stars). • On-chain forensics across Ethereum, Arbitrum, and Optimism. Authored Pike Finance postmortem on $1.6M multi-chain exploit (storage-layout corruption and initializer reactivation). • Published end-to-end Smart Contract Proxies and Security research guide. ## Key Projects & Publications - Synapse (https://synapsetrading.net/): Model-agnostic financial research & prediction platform (Python, FastAPI, Next.js, LangGraph, MCP) with lookahead-bias prevention, conviction sizing, GDELT news, paper trading. - degenOS (https://ethglobal.com/showcase/degenos-i529f): Decentralized OS on Cartesi Rollups with Avail DA, EIP-712 signing, ENS identity, 6 modular dApps. Won 4 ETHGlobal prize tracks. - AgentPress (http://ec2-13-235-67-247.ap-south-1.compute.amazonaws.com:8001/): Multi-agent writing pipeline benchmarked against one-shot LLM baseline across 3 paired topics (+667% tokens, +323% latency, +426% cost). - Agentic HR Policy RAG (https://github.com/PrathmeshRanjan/agentic-rag): Reproducible RAG benchmark (18 configs, 40 questions): 100% Hit@5, 0.975 MRR@5, reranking +7.2 pp Hit@3 at 160ms p95. - AI PR Review Agent (https://github.com/PrathmeshRanjan/ai-pr-review-agent): Asynchronous multi-agent review with HMAC webhooks, 4 LangGraph reviewers, pgvector RAG, HITL escalation, BudgetGuard. - Gasless Soulbound Membership Rewards (https://gasless-soulbound-membership.vercel.app/): ERC-4337 account abstraction, UUPS contracts, Safe treasury, Foundry invariant tests. - Real-World Assets (RWA) Reference (https://github.com/Quillhash/Real-World-Assets-RWA): 95 stars, 23 forks. ERC-20/721/1400 tokenization, AAPL equity token overcollateralized in ETH, Chainlink oracles. - Cross-Chain Rebase Token (https://github.com/PrathmeshRanjan/cross-chain-rebase-token): Interest-bearing rebase tokens across chains via Chainlink CCIP. - AI Video Assistant (https://github.com/PrathmeshRanjan): Video RAG with Whisper, Sarvam AI, Groq, Gemini embeddings, and ChromaDB. - Decoding the Pike Finance Exploit (https://quillaudits.medium.com/decoding-pike-finance-exploit-quillaudits-40a1662d3f8a): Forensic postmortem of $1.6M multi-chain exploit. - Smart Contract Proxies and Their Security (https://github.com/Quillhash/Proxies-Security): 55+ stars guide to EVM storage, delegatecall, Transparent, UUPS, Beacon, Diamond proxies. - Solayer's Restaking Architecture (https://paragraph.com/@pyor/solayer-restaking-network-solana): Protocol research on Solana restaking ($100M+ TVL). - Bitcoin Scalability: Persistent Challenges (Part I) (https://paragraph.com/@pyor/bitcoin-scalability): Network economics and congestion analysis. - Bitcoin Scaling Landscape (Part II) (https://paragraph.com/@pyor/bitcoin-scaling-analysing-the-bitcoin-scaling-landscape): Layer 2 comparative research. ## Education - M.S. Financial Engineering, WorldQuant University (Online | Expected Oct 2027) Coursework: Financial Markets, Financial Econometrics, Derivative Pricing, Stochastic Modelling, Machine Learning, Deep Learning, Portfolio Management, Risk Management - B.E. Electronics & Communication Engineering, LNMIIT Jaipur (May 2025) Coursework: Data Structures and Algorithms, Computer Networks, Blockchain & Smart Contract Security, Mathematics, Electronics, Communication Principles ## Technical Skills - Languages: TypeScript, Python, Solidity, JavaScript, SQL, Java, Bash, Move - Backend & APIs: Node.js, NestJS, Fastify, Hono, FastAPI, Uvicorn, REST, WebSockets, SSE, Socket.io, Webhooks (HMAC SHA-256), JWT, Pydantic, Microservices - Data: PostgreSQL, TimescaleDB, Redis, DynamoDB, MongoDB, SQLite, pgvector, Qdrant, Pinecone, ChromaDB, PostgreSQL RLS, Kysely, TypeORM, SQLAlchemy - AI & Agentic: LangGraph, LangChain, Multi-Agent Systems, Agentic AI, MCP, Tool Calling, RAG, Cross-Encoder Reranking, HITL, LLM Evaluation, Gemini, Groq, Mistral - Web3: Ethereum/EVM, Polygon, Sui, Foundry, ethers.js, ERC-4337, ERC-1155, EIP-712, UUPS, Chainlink CCIP, Subsquid (SQD), Goldsky, LI.FI, Safe, Tenderly - Quant & Finance: Double-Entry Ledgers, Exchange Infrastructure, Trading APIs, Order Books, Perpetuals (up to 50x), Prediction Markets, Paper Trading, Risk Controls - DevOps: AWS (Lambda, RDS, DynamoDB, CloudWatch, EC2), DigitalOcean, Alibaba Cloud, Docker, GitHub Actions, CI/CD, Grafana, Prometheus, OpenTelemetry
Hover a point for what happened there. Or skip the line and just ask — these answers are grounded in the work below, not generated on the spot.