Piyush Pandey — Senior Software Engineer

Senior Software Engineer & GenAI Explorer | 5.5+ years building distributed systems on Node.js, TypeScript & React — now actively applying LLMs, RAG pipelines, and agentic AI to real engineering problems. Currently tech-leading a 16-engineer team at Think Future Technologies, owning end-to-end architecture, client discovery, and delivery for a large-scale enterprise platform. What I've built: → AI Meeting Assistant (2026) — End-to-end GenAI tool using LangChain LCEL + LangGraph, Mistral AI, OpenAI Whisper, RAG with ChromaDB & HuggingFace embeddings. Handles multilingual transcription (EN/Hinglish), summarization, action-item extraction, and conversational Q&A over transcripts via vector search. → SimplifyPath — A pre-MCP-era personal-assistant platform (before Claude connectors existed) enabling employees to execute tasks across 15+ SaaS apps through a unified conversational interface. Org-wide production use at a leading Indian auto-marketplace (NDA, ~5K+ employees), sustaining 10K+ background jobs/day on Kafka, BullMQ, and Redis. → Enterprise Field-Ops Platform — Node.js, PostgreSQL, Kubernetes, Azure. 5-level RBAC hierarchy, real-time ERP sync, Prometheus/Grafana/Loki observability. 60+ features in 10 months, p95 latency cut from ~3.2s to ~600ms. → Botifier — No-code multi-channel bot builder running 50+ bots in parallel per tenant on XState finite state machines, with Zapier as a universal action layer (5,000+ downstream apps). Core strengths: - GenAI: LangChain, LangGraph, RAG, Vector Stores, LLM orchestration, Prompt Engineering - Backend: Node.js, TypeScript, Kafka, BullMQ, Redis, PostgreSQL, Microservices - Platform: Kubernetes, AWS, Azure, Docker, CI/CD, Prometheus/Grafana - Leadership: Tech lead for 16-engineer team, system design, client discovery Open to senior backend, full-stack, GenAI engineering, or tech-lead roles at product-led teams. 📧 piyush.pande94@gmail.com

Stackforce AI infers this person is a SaaS Backend Engineer with expertise in Generative AI and system architecture.

Location: Delhi, India

Experience: 5 yrs 5 mos

Skills

Backend
Leadership
Frontend

Career Highlights

Led a 16-engineer team to deliver complex features.
Built a GenAI tool for multilingual transcription and Q&A.
Architected scalable systems with real-time observability.

Work Experience

Think Future Technologies

Senior Software Engineer (1 yr 9 mos)

Software Engineer (1 yr 11 mos)

JungleWorks

Associate Software Engineer (1 yr 9 mos)

Education

Bachelor of Technology - BTech at Chandigarh Engg College, Landran

Piyush Pandey

Senior Software Engineer

Delhi, India5 yrs 5 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Led a 16-engineer team to deliver complex features.
Built a GenAI tool for multilingual transcription and Q&A.
Architected scalable systems with real-time observability.

Stackforce AI infers this person is a SaaS Backend Engineer with expertise in Generative AI and system architecture.

Contact

Skills

Core Skills

BackendLeadershipFrontend

Other Skills

Node.jsTypeScriptPostgreSQLKubernetesAzurePrometheusGrafanaLokiCI/CDReactXStateKafkaBullMQRedisMongoDB

About

Experience

5 yrs 5 mos

Total Experience

2 yrs 8 mos

Average Tenure

3 yrs 8 mos

Current Experience

Think future technologies

2 roles

Senior Software Engineer

Promoted

Sep 2024 – Present · 1 yr 9 mos · Remote

Tech-leading a 16-engineer team across backend, frontend, and QA on a large-scale enterprise field-operations platform (confidential client) serving 1K+ daily active users.
Own sprint planning (ClickUp), feature decomposition, PR reviews (~25/week), and release delivery
Shipped 60+ complex features in 10 months without a production rollback, sustaining ~6 releases/month
Architected a 5-level hierarchy and role-based access system (managers → zonal heads → area heads → field officers → distributors) spanning ~50K+ nodes in the org graph
Designed the backbone on Node.js, PostgreSQL, Kubernetes, and Azure — with real-time ERP sync, Azure Blob (signed URLs), and Prometheus/Grafana/Loki observability
Tuned PostgreSQL indexes on the reporting hot-path, cutting p95 latency from ~3.2s to ~600ms
Reduced deployment time from ~40min to ~12min through layer caching, parallel K8s rollouts, and smarter test selection
Own client-facing technical discovery: translate ambiguous business needs into flow diagrams, API contracts, and phased delivery plans
Tech: Node.js, TypeScript, PostgreSQL, Kubernetes, Azure (Blob, Cloud Services), Prometheus, Grafana, Loki, CI/CD

Node.jsTypeScriptPostgreSQLKubernetesAzurePrometheus+5

Software Engineer

Sep 2022 – Aug 2024 · 1 yr 11 mos · Remote

Architected and shipped two major production systems at TFT:
SimplifyPath — Cross-App Personal Assistant Platform (Jun 2023 – Dec 2024)
Built a pre-MCP-era assistant (before Claude connectors existed) executing tasks across 15+ SaaS apps through a unified conversational interface
Currently in org-wide use at a leading Indian auto-marketplace (NDA), ~5K+ employees
Designed the async backbone on Kafka, BullMQ, and Redis — sustaining 10K+ jobs/day with <1% failure rate and automatic retry with exponential backoff
Built 15+ pluggable app-connector modules abstracting third-party APIs (Slack, Gmail, calendars, internal SaaS) behind a common contract; new integrations ship in ~1 day
Cut hot-path p95 latency 3–5x (~2.4s → ~600ms) by parallelizing waterfall calls and offloading work to async queues
Led frontend code reviews and React component architecture
Botifier — No-Code Multi-Channel Bot Builder (Nov 2022 – Jun 2023)
Co-designed a drag-and-drop bot builder running 50+ bots in parallel per tenant across WhatsApp, Telegram, and in-app channels
Built the runtime on XState finite state machines — each bot runs as an isolated interpreter instance, hot-swappable at runtime with zero session disruption on flow updates
Integrated Zapier as a universal action layer, unlocking 5,000+ downstream apps through a dynamic input/output pipeline
Authored an internal SDK encapsulating REST APIs, state-machine triggers, and DB interactions — reused verbatim to ship the bilingual (EN/HE) Experton product in ~3 months from scratch
Deployed on AWS (EC2, S3, CloudWatch) with CI/CD
Tech: Node.js, TypeScript, React, XState, Kafka, BullMQ, Redis, MongoDB, AWS (EC2, S3, CloudWatch), Docker

Node.jsTypeScriptReactXStateKafkaBullMQ+6

Jungleworks

Associate Software Engineer

Nov 2020 – Aug 2022 · 1 yr 9 mos · Remote

Built and maintained the integrations layer for Tookan — a delivery SaaS handling 1M–10M orders and ~500K MAU across food, grocery, and healthcare verticals.
Owned a microservice containing 50+ payment gateway flows powering multiple products across the Jungleworks ecosystem
Integrated 5+ payment gateways (Stripe Connect/Hold, Razorpay, Paytm, Paystack) behind a unified SDK interface — adding a new provider became a ~2-day change instead of a multi-week refactor touching every caller
Integrated 4+ SMS gateways (Twilio, Plivo, MSG91, Releans) for real-time delivery notifications
Published the Tookan app on the Shopify App Store, facilitating seamless order sync for 500+ merchants
Integrated marketplaces via webhooks (WooCommerce, Zoho Commerce, Petpooja), reducing manual order handling by 40%
Designed an intelligent fallback (F2) routing flow auto-dispatching orders to Dunzo and DHL when internal agents are saturated — improving fulfillment rates by 30%
Built scalable web applications on Node.js and MySQL; managed production servers on AWS EC2 with HAProxy and PM2
Mentored junior developers on coding standards, review workflows, and integration patterns
Tech: Node.js, JavaScript, MySQL, AWS EC2, HAProxy, PM2, REST APIs, Webhooks