About Me

AI Engineer and Full-Stack Developer with 4+ years of experience building production web applications, agentic AI systems, and intelligent automation pipelines. Specialized in LLM application development including multi-tool agentic orchestration, RAG-based knowledge retrieval using pgvector, structured output pipelines with forced tool use and JSON schema validation, multi-provider AI integration (Anthropic Claude, OpenAI, Google Gemini, xAI Grok), and streaming AI systems with SSE. Strong UI/UX sensibility applied to both consumer-facing products and human-in-the-loop AI interfaces. Experienced with workflow automation using Make.com for trigger-based AI pipelines. Equally strong on the full-stack side — building consumer-facing UIs in React and TypeScript, REST API design, real-time systems, payment pipelines, and cloud deployment on Vercel and AWS.

4+ years building agentic AI systems, LLM applications, and production web applications

Specialized in multi-tool orchestration, RAG with pgvector, and streaming AI systems (SSE)

Multi-provider AI integration: Anthropic Claude, OpenAI GPT-4o, Google Gemini, xAI Grok

Full-stack proficiency: React, Next.js, TypeScript, Node.js, Python, FastAPI

Production experience with Stripe Connect, real-time systems, and workflow automation

Cloud deployment expertise: Vercel, AWS (Lambda, EC2, RDS, CloudWatch)

Work Experience

AI Engineer

Whispering AI

Oct 2024 – Present
Scottsdale, Arizona, United States
  • Contributed to architecting and deploying a production agentic coding assistant that reads and writes files directly to GitHub repositories via a two-phase planning/execution loop, with user-configurable model selection (Claude Sonnet 4, Opus, Haiku), temperature control, draft/ship modes, and per-request token cost tracking
  • Contributed to building a multi-tool agentic chatbot with a 5-iteration tool-call loop, persistent session history, and custom tools — knowledgebase search, project portfolio search, and contact form triggering — deployed as a Supabase Edge Function with RAG-style retrieval using pgvector-backed JSON knowledge stores as the truth source
  • Contributed to engineering an AI Assessment Report Generator using forced tool use (tool_choice) and server-side JSON schema validation to guarantee structured output, producing personalized reports with graded observations, prioritized recommendations, and ROI estimates
  • Contributed to developing a streaming internal assistant with real-time SSE delivery, injecting live CRM data — milestones, backlog items, customers, sales leads, and team members — into context per request using a configurable AI directives system with drag-and-drop prompt management UI
  • Contributed to building a multi-provider AI content generation system supporting Anthropic Claude, OpenAI GPT-4o-mini, and Google Gemini — with per-customer model, temperature, max_tokens, and top_p configuration, brand document injection, and email template context for HTML email body generation
  • Contributed to implementing an AI-powered report generator using Claude with Gemini fallback, exponential backoff retry logic on API failures, and industry-specific context injection from a database-backed reference document system
  • Contributed to designing and building an AI scaffold generator that takes a product intake form and outputs a complete project scaffold — tech stack, npm packages, PostgreSQL schema with RLS, API setup checklist, and sequenced build prompts — as structured JSON via forced tool use

Full-Stack Developer

CareerManzil

Jun 2025 – Present
Fremont, California, United States
  • Built responsive, consumer-facing web applications using React, TypeScript, and Vite, translating design specs into pixel-accurate, cross-device UI
  • Developed and maintained reusable component libraries using Tailwind CSS and Shadcn/UI, establishing consistent design system patterns across products
  • Integrated REST APIs into dynamic, data-driven UI workflows including JWT-based authentication, protected routes, and stateful user sessions
  • Collaborated closely with designers and product stakeholders to translate wireframes and design specifications into responsive, production-ready interfaces, ensuring pixel-accurate implementation across breakpoints
  • Contributed to performance optimization initiatives including lazy loading, code splitting, and bundle size reduction, improving page load times and Core Web Vitals scores across key user flows

Software Engineer – Platform & Reliability

Comprobase Inc.

May 2024 – Jun 2025
Dallas, Texas, United States
  • Owned observability and operational reliability for AWS-hosted customer-facing services, maintaining 99.99% uptime across EC2, RDS, CloudFront, and S3
  • Built Python and Node.js automation for monitoring, alerting, and incident response pipelines, reducing mean time to resolution by 30%
  • Authored technical runbooks and internal documentation improving Tier-1 resolution rates by 25%

AI Solutions Engineer & Advisor

FewCuts

Jan 2020 – May 2024
Remote
  • Contributed to building a viral moment detection pipeline using xAI Grok-4 with chain-of-thought reasoning, genre-specific prompt templates per content type, and sentence-level timestamp injection — producing confidence-scored clip candidates with separate reasoning token tracking
  • Contributed to a speech-to-text transcription microservice using Groq Whisper (whisper-large-v3-turbo) with MD5-based disk caching to prevent redundant API calls, producing word-level timestamped transcripts for downstream processing
  • Contributed to integrating pyannoteAI speaker diarization via async polling pipeline, attributing speech segments to individual speakers for multi-speaker caption rendering
  • Contributed to architecting a Python FastAPI microservice pipeline for end-to-end video processing — transcription, diarization, viral detection, and TikTok-style caption burning via FFmpeg and ASS subtitle generation

Technical Skills

AI & LLMs

Anthropic Claude API
OpenAI API (GPT-4o, gpt-image-1, Whisper)
Google Gemini
xAI Grok-4
Groq

AI Engineering

Agentic workflows
Multi-tool orchestration
Forced tool use
RAG (pgvector)
Context injection
Prompt engineering
Structured output
Streaming (SSE)
Langchain
Multi-provider fallback
Token cost tracking
Make.com & n8n

Speech & Media AI

Groq Whisper (speech-to-text)
pyannoteAI (speaker diarization)
FFmpeg video processing

Frontend

React
Next.js
TypeScript
JavaScript
Vite
Tailwind CSS
HTML5
CSS3
Framer Motion

UI & Components

Shadcn/UI
Radix UI
Recharts
Reusable component libraries
D3.js
Design systems
WCAG 2.1 AA

Backend

Node.js
Python
FastAPI
Deno
REST APIs
Supabase Edge Functions
JWT
OAuth2

Databases

PostgreSQL
Supabase
MySQL
MongoDB
PostGIS
Row Level Security
pgvector

Cloud & DevOps

Vercel
AWS (EC2, Lambda, RDS, CloudWatch, Lightsail, S3)
CI/CD pipelines
Docker

Payments

Stripe Connect
PaymentIntents
Webhooks
Subscription lifecycle

Education

Master of Science in Computer Science

University of Texas at Tyler

Aug 2022 – May 2024
Tyler, Texas, United States

Bachelor of Engineering in Computer Science

Osmania University

Jul 2017 – Jun 2021
Hyderabad, India

Certifications & Awards

Certifications

Microsoft Azure Fundamentals

Cyber-Data Foundations

University of Texas at Tyler

AWS Certified Cloud Practitioner

In Progress

Awards

₹100,000 R&D Grant — SoundTrap Project

Hyderabad City Police & Telangana State Government

2022