~/abdullah $ whoami — lead full stack engineer

Systems that
hold up at
3 a.m.

I build and scale web platforms, distributed microservices and event-driven systems in TypeScript, React, Next.js, Node.js and AWS — plus AI automation and agentic AI services in Python and FastAPI, and Solidity smart contracts across Ethereum and Solana. Currently leading 4–6 engineers on real-time AI and Web3 products used across global contact centers.

See the work→$Get in touch

github.com/abdullah-safdar linkedin.com/in/abdullah-baig

typescriptreact · next.jsnode.js · nestjspython · fastapiagentic aiai automationevent-driven microservicesrabbitmqaws lambdapostgresqlmongodbredisgraphql · grpcllm integrationsolidity · smart contractsethereum · solanadocker · ci/cdteam leadershiptypescriptreact · next.jsnode.js · nestjspython · fastapiagentic aiai automationevent-driven microservicesrabbitmqaws lambdapostgresqlmongodbredisgraphql · grpcllm integrationsolidity · smart contractsethereum · solanadocker · ci/cdteam leadership

01experience7yearsBuilding and scaling production web platforms

02leadership4–6engineersLed end-to-end — architecture through on-call

03performance25–30% p95Latency cut by re-architecting to event-driven services

04velocity30–40% fasterFeature delivery via AI-assisted development workflows

selected work

shipped 2019 — 2026

~/work/phron-ai202601Phron AIprompt─▶llm gen─▶compile─▶on-chainWeb3 AI compiler platform using LLMs to generate and compile Solidity and Rust contracts, with MetaMask on-chain auth and usage-based payments.LLM-generated contracts compiled in-browser, wallet-native authLLMSolidityNext.jsread case→~/work/call-agent202602Call Agentaudio in─▶stt─▶llm stream─▶tts outReal-time AI voice agent on streaming LLM responses with low-latency text-to-speech — built for global contact-center load.Sub-second response on streaming LLM + TTS pipelineRealtimeDeepgramWebSocketsread case→~/work/launchpad202503Token Launchpadconfig─▶sign─▶deploy─▶verifyAutomated Solana bot for Pump.fun launches: wallet interactions, transaction signing and on-chain deployment workflows.Automated signing and on-chain deploys, zero manual stepsSolanaNode.jsWeb3read case→~/work/agents202504Agentic Automationstask─▶plan─▶tool calls─▶approvalAgentic AI workflows that call internal tools and APIs on their own — task planning, retries and human-in-the-loop approvals.Python + FastAPI services orchestrating multi-step agentsPythonFastAPIAgentsread case→~/work/workflows202405Workflow Builderdrag─▶validate─▶publish─▶tenantDrag-and-drop workflow builder and admin surface for platform configuration across multi-tenant accounts.Multi-tenant config surface with role-based accessNext.jsRBACDesign systemread case→~/work/propskit202206Propskitquery─▶resolver─▶batch─▶renderRebuilt the data layer on GraphQL to kill over-fetching, with measurable gains in API flexibility and frontend performance.Over-fetching removed — faster frontends on the same APIGraphQLPerfReactread case→

capabilities

01Frontend architectureReact and Next.js at scale — layered UI, performance budgets, design systems.

02Backend & APIsNode.js, NestJS, Python and FastAPI across REST, GraphQL and gRPC.

03Distributed systemsEvent-driven services on RabbitMQ and Redis, idempotent and retry-safe.

04Cloud & DevOpsAWS Lambda, Step Functions, SQS, Docker, CI/CD and observability.

05AI automation & agentsAgentic workflows, LLM integration, real-time voice and tool calling.

06Web3 & smart contractsSolidity on Ethereum, Solana programs, wallet auth and on-chain payments.

Languages

TypeScript · JavaScript · Solidity · Python · PHP · SQL

Frontend

React · Next.js · UI architecture · web performance · responsive design

Backend

Node.js · NestJS · REST · GraphQL · OpenAPI · gRPC · microservices · event-driven architecture · RabbitMQ · WebSockets

Security & auth

JWT · OAuth 2.0 · RBAC · API rate limiting · secure session management

Cloud & DevOps

AWS Lambda, Step Functions, Amplify, SQS, S3 · Docker · CI/CD · monitoring & observability

Data & storage

PostgreSQL · MongoDB · DynamoDB · Redis · Elasticsearch · database modeling

AI-assisted development

Claude Code · Cursor · ChatGPT · Copilot workflows · AI-driven review, test generation, refactoring

Leadership

Team leadership · mentorship · performance management · incident management · delivery ownership

experience

Lead Software Engineer

Plutus Software House

2022 — Present

—Lead 4–6 engineers building real-time AI and Web3 platforms used across global contact centers.

—Re-architected the core backend into event-driven microservices (Node.js, RabbitMQ, Redis) — p95 latency down 25–30% while absorbing several-fold traffic growth without proportional cost.

—Drove team-wide adoption of AI-assisted development, automating boilerplate, scaffolding and test generation to cut delivery time an estimated 30–40%.

—Led a zero-downtime MongoDB → PostgreSQL migration with dual writes and background backfill — no customer-facing interruption.

Full Stack Engineer

Scrapespot

2021 — 2022

—Developed backend services in Node.js and NestJS across REST and GraphQL, favouring modular, maintainable architectures.

—Integrated GraphQL data layers to reduce over-fetching and streamline frontend–backend communication.

Software Developer

Bargo Ventures

2019 — 2021

—Designed and implemented RESTful APIs with Node.js and Laravel supporting scalable backend systems.

—Improved frontend UI/UX and performance alongside designers on responsive interfaces.

writing

notes on architecture, AI-assisted delivery & web3

Event-driven without the chaosAug 14, 2026

Dedup keys, idempotency and circuit breakers — the three pieces that turned a flaky integration layer into one I stopped getting paged about.

Zero-downtime migration, MongoDB to PostgresJun 09, 2026

Dual writes, a background backfill, and the read-path switch. What I'd do differently on the next one.

Shipping 30% faster with AI in the loopApr 06, 2026

Where Claude Code and Cursor genuinely pay off on a team of six — and the review habits that keep quality from sliding.

Streaming LLMs into a phone callFeb 11, 2026

Latency budgets for a voice agent: token streaming, TTS chunking, and the barge-in problem nobody warns you about.

All posts

~/abdullah $ hire

Open to lead & staff-level backend and full-stack work.