I find the real workflow.
I map the people, data, handoffs, and edge cases before deciding what should be automated.
Open to the right Full-Stack AI opportunity
NIKHIL RAJPUT · FULL-STACK AI ENGINEER
I build AI products that move from conversation to production.
I turn business requirements into reliable agents, automation systems, full-stack platforms, data pipelines, and SaaS products—then help them survive real users, real integrations, and real operational constraints.
2+ years hands-onMadhya Pradesh, IndiaEnd-to-end ownership
Agentic AI
Backend systems
Automation
SaaS products
2+ years · international client deliveryMove your cursor
From ambiguityRequirements → architecture
Across the productFrontend → backend → AI
Through deliveryTesting → deployment → iteration
WHAT I BRING
I connect the intelligence layer to the whole product.
My strongest work happens where AI, full-stack engineering, automation, integrations, and product decisions all meet.
I’m a Full-Stack AI Engineer based in Madhya Pradesh, India, with 2+ years of hands-on experience building AI-powered products, automation systems, full-stack SaaS platforms, data pipelines, and production integrations for international clients.
I work from discovery through architecture, development, testing, deployment, and production iteration—with practical AI automation and reliable full-stack systems at the center.
See the full picture on Labs ↗I own the connected build.
Interfaces, APIs, databases, agents, workers, integrations, testing, and deployment remain one engineering problem.
I design for control.
Structured outputs, guardrails, bounded tools, fallbacks, observability, and human handoff are part of the architecture.
WHAT I'VE SHIPPED
Production systems with the full engineering story.
Client delivery across AI, automation, data acquisition, and SaaS—followed by focused personal products and additional builds for wider technical proof.
Multi-Tenant AI Property Management Platform
One connected system for property data, tenant operations, communication, payments, and service requests.
A full-stack, multi-tenant platform that combines property acquisition, AI-assisted operations, action-oriented WhatsApp experiences, and workflow automation.
Node.jsExpress.jsNext.jsPostgreSQLChromaDBLangChainRedisPlaywrightn8nWhatsAppPeach PaymentsDockerBitbucket
Built Playwright acquisition pipelines for property listings and connected LangChain RAG with ChromaDB, removing repetitive data entry and enabling grounded tenant interactions.
Created WhatsApp agents for tenant queries, maintenance requests, payment reminders, and lease enquiries, with controlled actions against live backend data.
Automated outreach and social publishing across three channels with n8n; integrated Redis, Peach Payments, Docker, and repeatable Bitbucket-based delivery workflows.
AI Intake & Lead Qualification System
An SMS intake workflow where deterministic rules and people stay in control.
The system automates information collection, qualification, follow-up, and escalation while keeping sensitive decisions inside explicit safety boundaries.
PythonFastAPITwilioLLM APIsPostgreSQLAWS App RunnerCloudFrontSQSTerraform
Built configurable conversation flows with structured fact extraction, weighted scoring, deterministic qualification rules, and target-based handoff to human operators.
Applied crisis detection, restricted-response rules, output validation, and fallbacks across four claim categories to prevent unauthorised advice while preserving natural conversation.
Deployed production services on AWS App Runner with SQS-based asynchronous processing and a CloudFront-delivered operator dashboard.
Conversational Voice AI & Hiring Automation
Real-time voice automation with multilingual understanding and a clear human exit.
A live calling system that maintains context, collects information, understands intent, and moves people to WhatsApp or human support when required.
PythonFastAPIPlivoLLMsSTT/TTSElevenLabsSarvamWhatsAppHubSpotNext.jsDockerGCP Cloud Run
Orchestrated bidirectional WebSocket audio, energy-based VAD, STT/TTS, and five-provider LLM fallback for responsive conversations and resilient failover.
Integrated Hindi and English voice processing with WhatsApp post-call workflows to automate daily outbound calls and reduce repetitive screening work.
Built a real-time operations dashboard for call outcomes, qualification, callback scheduling, and human follow-up.
Development Application Tracking Platform
Council records and planning documents become enriched, actionable CRM leads.
A background acquisition and enrichment platform that monitors public portals, extracts planning data, enriches records, and routes qualified opportunities.
PythonPlaywrightAWS BedrockAWS TextractReact.jsPostgreSQLMonday.comABR APIRP DataEC2S3Docker
Deployed Playwright workers across 20+ council portals to capture new development applications and linked planning documents on a scheduled basis.
Applied AWS Bedrock and Textract to unstructured PDFs, then enriched records through ABR and RP Data APIs with contact and property details.
Built a React monitoring dashboard with automated Monday.com routing, scraper-health visibility, and traceable lead delivery.
WORKFLOW AUTOMATION
Automation that removes handoffs, not control.
01 CONNECTED SUITE
Outreach, email, invoice processing, CRM updates, and notifications—joined through APIs, rules, and measured AI assistance.
Business Workflow Automation
Deterministic orchestration for repeatable work, with LLM reasoning only where input becomes unstructured.A connected automation toolkit for outreach, email campaigns, invoice processing, Sheets, CRM updates, and operational notifications.
n8nChrome ExtensionsNode.jsBrevo APIGoogle SheetsAPIsWebhooksEmailLinkedInCRM APIsLLMs
LinkedIn outreach
Built a Chrome extension and backend for profile-aware messages, queuing, sequenced follow-ups, controlled delays, and A/B variants.
Email campaigns
Connected Brevo and Google Sheets for personalised campaigns with merge fields, delivery tracking, bounce handling, drip sequencing, and analytics.
n8n orchestration
Designed an invoice pipeline across webhook, Gmail, and Drive sources, joining extraction, validation, retries, and downstream updates through APIs and conditional routing.
PERSONAL PROJECTS
AI products I’m shaping end to end.
03 AI PRODUCTS
Independent product work where architecture, evaluation, data boundaries, asynchronous systems, and user experience evolve together.
WebCopilot AI
Natural-language web workflows that plan, execute, self-heal, and become faster after approval.An enterprise-oriented autonomous web agent and RPA platform that turns a user’s objective into a reviewed, versioned, and repeatable browser automation workflow.
- Built a dual execution engine: a compiled Playwright fast path for approved repeat runs, plus a self-healing ReAct path that uses live DOM and screenshot evidence when the workflow changes.
- Implemented an approval-first lifecycle covering natural-language intent, plan and extraction schema generation, human review, immutable agent versions, execution history, and safe rollback.
- Structured the product as separate Next.js, NestJS, worker, notifier, and demo services; currently hardening deployment with Google Cloud services, managed secrets, queues, storage, and production observability.
Next.js 16TypeScriptNestJSPlaywrightPrismaPostgreSQLGeminiResendDocker
AI Career & Job Intelligence Platform
A modular AI-first operating system for the complete job-search journey.Career profiles, resumes, portfolios, job intelligence, matching, applications, and interview preparation live inside one connected product architecture.
- Designing scheduled connectors across 5+ job platforms with parsing, normalisation, duplicate detection, skill extraction, embeddings, and vector search.
- Architecting Redis and Taskiq workers for asynchronous parsing, scraping, matching, resume generation, and notifications without blocking the product experience.
- Building LangGraph and LiteLLM workflows across OpenAI, Gemini, Claude, and Groq with LangSmith observability and voice-powered interview simulation.
Next.js 16FastAPIPostgreSQLWeaviateRedisTaskiqLangGraphLiteLLMLangSmithOpenAIGeminiClaudeGroqDocker
Personal Finance AI Agent
Statements become structured transactions, useful analytics, and grounded conversation.A finance platform for bank-statement ingestion, categorisation, spending analytics, budget tracking, and source-grounded natural-language analysis.
- Combined deterministic processing with RAG and pgvector to answer grounded questions over categorised transaction data.
- Built an extensible PDF extraction pipeline for Indian bank formats, including HDFC, SBI, and IDFC, with validation and automated categorisation.
- Developed spending visualisations, budget tracking, and anomaly detection to surface unusual transactions.
PythonFastAPINext.jsPostgreSQLpgvectorRAGLLM APIsDocker
ADDITIONAL PRODUCTS
More proof across product categories.
05 MORE SYSTEMS
A compact view of the wider full-stack, integration, mobile, language, and geospatial systems I’ve worked on.
Omnichannel Social & Advertising Automation Platform
Full-stack publishing, moderation, and campaign operations across major social platforms.A centralised product for secure channel connections, personalised publishing, moderation, and advertising workflows.
- Implemented OAuth 2.0 across Meta, TikTok, YouTube, LinkedIn, and X with secure token lifecycle management.
- Built a multi-platform publishing engine with scheduling, AI-assisted optimisation, campaign analytics, and production-ready integrations.
Next.jsReact.jsNode.jsExpress.jsPostgreSQLOAuth 2.0MetaTikTokYouTubeLinkedInX
Educational ERP
Admissions, attendance, placements, payments, and communication in one operational system.A full-stack ERP built around the daily workflows of students, staff, and administrators.
- Built QR and GPS attendance workflows with automated WhatsApp notifications for students and administrators.
- Integrated PhonePe fee collection and receipt generation with backend APIs across six academic workflows.
React.jsNode.jsExpress.jsMySQLWhatsApp TemplatesPhonePeQRGPS
Real Estate CRM Bridge
A cross-platform bridge between third-party real estate systems and custom web and mobile tools.Property listings, contacts, and transactions stay synchronised across an external CRM, operational dashboards, and a mobile app.
- Built a FastAPI integration layer with real-time Firebase updates for React and React Native clients.
- Delivered an Expo mobile workflow for on-the-go property management, client communication, and listing updates.
PythonFastAPIReactReact NativeExpoFirebaseREST APIs
Farmer Interview Transcription Pipeline
Hindi field interviews become schema-enforced agricultural records in minutes.An audio-to-structured-data pipeline that converts long-form farmer conversations into data ready for downstream analysis.
- Combined Google Cloud Speech-to-Text and Gemini to turn long Hindi field interviews into structured agricultural records.
- Classified crop, soil, and farming-practice fields using Pydantic-enforced schemas with retry and review paths.
PythonFastAPIGoogle Cloud STTGeminiPydantic
Precision Agriculture Platform
Satellite-powered farm analysis with layered vegetation, soil, and crop-health intelligence.A geospatial platform that transforms remote-sensing data into zone-wise farm recommendations and temporal analysis.
- Built a nine-layer analysis system for NDVI, soil moisture, crop health, and NPK trends using Sentinel Hub and ISRO VEDAS imagery.
- Developed an OpenLayers workspace for farm boundaries, time-series analysis, and prescription maps.
PythonFastAPIReactOpenLayersSentinel Hub APIISRO VEDASPostgreSQL
HOW I WORK
Useful AI begins before the model call.
I treat the model as one part of a production system. The workflow, data, controls, interfaces, and recovery paths decide whether the product is genuinely useful.
Understand the operation
I start with the people, data, handoffs, edge cases, and business outcome before choosing a model or framework.
Separate rules from reasoning
I define what must stay deterministic, where AI adds value, which tools are allowed, and when a person must take over.
Build the complete path
I connect interfaces, APIs, agents, workers, databases, integrations, observability, and deployment as one system.
Test it in the real workflow
I validate normal and failure paths, ship incrementally, watch production behavior, and improve the system with users.
MY TECHNICAL RANGE
Tools grouped by the part of the system they strengthen.
I choose the stack around the workflow, deployment environment, integration surface, and reliability needs—not around a fixed template.
AI & GenAI
LLM APIsAI AgentsRAGStructured OutputsEmbeddingsVector SearchGuardrailsTool CallingLangChainLangGraphLangSmithLiteLLMOpenAIGeminiAnthropicAWS Bedrock
Backend
PythonFastAPINode.jsExpress.jsREST APIsWebSocketsWebhooksAsync ProcessingBackground WorkersAuthentication
Frontend
React.jsNext.jsTypeScriptJavaScriptReact NativeTailwind CSSResponsive DashboardsOpenLayers
Automation
n8nPlaywrightBrowser AutomationWorkflow OrchestrationWhatsApp AutomationVoice AITwilioData PipelinesWeb Scraping
Data
PostgreSQLMySQLMongoDBChromaDBWeaviatepgvectorRedisSQLAlchemy
Cloud & DevOps
AWS EC2S3App RunnerCloudFrontSQSLambdaTextractGCP Cloud RunDockerGitHub ActionsCI/CDTerraformBitbucketVercel
Integrations
HubSpotMonday.comMetaTikTokYouTubeLinkedInXWhatsAppPayment GatewaysABR APIRP DataBrevoFirebaseSentinel Hub
HOW MY RESPONSIBILITY GREW
Quality first. Then software. Now complete AI systems.
My progression explains how I work today: I still think about edge cases like QA, build systems like a software engineer, and connect AI to the entire product.
Singaji Software SolutionsJuly 2024 - Present
Progressive responsibility across product quality, full-stack delivery, and production AI automation.
01
Nov 2025 - Present
Full-Stack AI Engineer
I lead the development of AI-powered automation and full-stack systems across architecture, LLM workflows, backend services, integrations, testing, deployment, and production iteration.
- LLM applications, AI agents, RAG pipelines, structured extraction, tool calling, guardrails, and human-in-the-loop workflows
- Production systems with Python, FastAPI, Node.js, PostgreSQL, n8n, Playwright, Redis, AWS, and third-party APIs
02
Nov 2024 - Nov 2025
Software Engineer
I developed backend and full-stack applications for international clients across Australia, South Africa, and other markets.
- Node.js, Express.js, Python, FastAPI, React, Next.js, PostgreSQL, and MySQL
- REST APIs, webhooks, integrations, dashboards, scraping pipelines, data workflows, deployment, and production support
03
Jul 2024 - Oct 2024
QA Tester + Developer
I started close to product quality: testing real workflows while contributing to development, debugging, and application reliability.
- Functional, API, regression, and workflow testing
- Defect reproduction, edge-case validation, fix verification, and collaboration with developers
WHERE I FIT NEXT
Give me a real workflow, not an isolated AI demo.
I’m looking for a team where engineering ownership, AI capability, and business context belong in the same room.
Agents that can understand, decide, and act safely
Full-Stack AI Engineer · GenAI Engineer · AI Automation Engineer
Complete products around the AI experience
Full-Stack Developer · AI Product Engineer · Applied AI Engineer
From business friction to a shipped product
AI Solutions Engineer · Automation Engineer · Product-minded Engineer
WHAT A TEAM GETS
I don't stop at “the AI part.”
I can speak with the people using the workflow, turn ambiguity into decisions, and keep engineering aligned with the outcome.
I can move from system boundaries and database design into APIs, agents, dashboards, automation, integrations, and deployment.
I know when to use deterministic logic, where AI adds leverage, and where validation, fallback, observability, or human control is required.
START A CONVERSATION
Ready to make a real workflow smarter?
I'm based in Madhya Pradesh, India, and open to Full-Stack AI, GenAI, full-stack development, automation, and AI solutions roles. Tell me what your team is trying to make easier, faster, or more reliable.