Divansh Bajaj profile photo 1 on Tal

Divansh Bajaj

Lead Engineer | Ex - SDE-II @Anuvrittiksha | AWS Certified Developer | Full-Stack, Cloud & Serverless Architect | AI Agents • Vertex AI • Google ADK • Agentic Workflows

Bengaluru

2+ years of work experience

Chitragupta Urban Tech
Aug 20263 mosPresent

Bengaluru

• Lead a team of 10 engineers building B2G smart-city products across road infrastructure, waste management, and property trade domains - owning technical direction, sprint delivery, and client-facing outcomes. • Delivered rapid prototypes and PoCs on tight deadlines using AI-assisted development (Claude), converting business/client requirements into working demos that secured 10+ client approvals/contracts. • Partnered directly with business teams and government stakeholders to translate requirements into technical specs; synced with audit teams on security compliance and audit-log management for government-grade systems. • Drove code review standards and code quality improvements across the team, reducing production defects and release friction. • Architected UI layers surfacing ML model predictions on interactive geospatial dashboards (Kepler.gl, Google Maps) for city-scale infrastructure planning.
Chitragupta Urban Tech
Feb 20269 mosPresent

Bengaluru

- Architected and delivered scalable, production-grade web applications using Node.js, React.js, MongoDB, and PostgreSQL, containerized via Docker and deployed on GCP - serving 10k+ users at high request volume. - Built automated ETL/data sync pipelines using cron jobs and batch processing to handle simultaneous background operations, ensuring data consistency and reducing manual intervention across services. - Implemented Redis caching and optimized PostgreSQL query performance through proper indexing strategies and query restructuring, reducing API response latency by ~20–30%. - Established CI/CD pipelines using GitHub Actions and GitHub Packages with integrated security checks, enforcing deployment best practices and reducing release friction across the team.
Anuvrittiksha Edtech
Dec 20241 yr 2 mosJan 2026

Remote

- HowToAlgo v1.1 – Performance & Scale: Improved performance 40% (v1.0 → v1.1) via lazy loading and optimized DynamoDB GSIs; lowered p95 latency/TTI; added event-driven flows with DynamoDB Streams and ETL via AWS Glue; scaled to thousands of users. - HowToAlgo v1.2 – Integrity, Cost & Security: Implemented ScaNN plagiarism detection using Bedrock Titan embeddings + vector search using FAISS (replaced O(N²) with top-k); used Gemini Flash for solution clustering, anomaly detection, and auto-summaries; reduced DynamoDB cost 30% by auditing access patterns and right-sizing/adding GSIs; strengthened SEO by improving keywords (+40% ranking boost) & implemented robots.txt + sitemap.xml for better crawlability. - Leadership: Mentored 5 developers; led Agile sprints & code reviews, improving delivery quality and consistency.
Anuvrittiksha Edtech
Jun 20246 mosNov 2024

WFH

HowToAlgo v1.0 – Launch: Led design & launch; frontend Next.js/TypeScript/Tailwind, backend Node.js/Express/REST; AWS stack (API Gateway, Lambda, Cognito, DynamoDB); CI/CD with GitHub Actions; Judge0 + secure payment gateways; CloudWatch monitoring; resolved 100% of Critical/High defects pre-go-live. fixed 100% SAST findings and improved security with CSP and stricter headers.
Omninos Solutions
Oct 20238 mosMay 2024

Mohali district, Punjab, India

- Live Streaming App – Built a live-streaming platform (React.js, Node.js, MongoDB) using Agora for low-latency video, admin/agency panels, and OTP via Fast2SMS; applied solid system design; shipped via CI/CD to Hostinger/Ultahost. - Chat App – Developed a real-time chat app with Socket.IO/WebSockets and Firebase Realtime Database: 1:1 & group chats, presence/typing indicators, delivery/read receipts, and secure auth.
Solitaire Infosys Inc
Jul 20222 mosAug 2022

Mohali

Divansh’s AI Stack

antigravity
antigravity
claude code
claude code
claude
claude
github copilot
github copilot
hugging face
hugging face
openai
openai

Education

BITS Pilani Work Integrated Learning Programmes

Master's Degree, Artificial Intelligence and Machine Learning

BITS Pilani Work Integrated Learning Programmes

20262 yrs2028
Chandigarh Group of Colleges, Landran, Punjab

Bachelor of Technology - BTech, Information Technology

Chandigarh Group of Colleges, Landran, Punjab

20204 yrs2024

Join Divansh and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Divansh Bajaj on Tal

Headline

Lead Engineer | Ex - SDE-II @Anuvrittiksha | AWS Certified Developer | Full-Stack, Cloud & Serverless Architect | AI Agents • Vertex AI • Google ADK • Agentic Workflows

About

Senior Full-Stack Engineer & AI Architect with a passion for building autonomous, scalable systems that solve complex real-world problems. Currently pursuing an M.Tech in AI & ML from BITS Pilani (WILP), I specialize in combining robust cloud-native architectures with Generative AI and Agentic workflows. 🚀 Current Focus at Chitragupta Urban Tech: I lead the architecture of production-grade web applications serving 10k+ users. My work focuses on high-performance infrastructure, where I've successfully: - Reduced API latency by 20-30% through advanced PostgreSQL indexing and Redis caching strategies. - Architected automated ETL pipelines and event-driven systems to handle massive data volumes with zero manual intervention. - Established rigorous CI/CD standards and security protocols to streamline delivery for the engineering team. 💡 Innovation in AI & EdTech: As a core contributor to HowToAlgo, I evolved the platform from a launch-ready MVP to a sophisticated AI-driven ecosystem: - Agentic AI: Built automated content pipelines using Google ADK and Gemini, deploying agents via ECR/Fargate. - Vector Search: Implemented high-speed plagiarism detection using FAISS/Chroma and RAG-based architectures. - Cloud Optimization: Audited AWS access patterns to reduce infrastructure costs by 30% while improving performance by 40%. 🛠 Technical Arsenal: - Languages: TypeScript, JavaScript, Python, C++. - Frontend & Backend: Next.js, React, Node.js, Express, FastAPI. - Databases: PostgreSQL, MongoDB. - AI/ML: LLMs (Gemini, Bedrock), RAG, Vector Databases, Vertex AI, Google ADK. - Cloud & DevOps: AWS Certified Developer Associate (Lambda, DynamoDB, API Gateway, SQS), Docker, GCP, GitHub Actions. I thrive at the intersection of clean code and intelligent automation. Whether it's mentoring a team of developers or fine-tuning a vector search algorithm, my goal is to deliver measurable value and production-grade reliability.

Location

Bengaluru

AI stack

antigravity

claude code

claude

github copilot

hugging face

openai

Work experience

Software Development Lead at Chitragupta Urban Tech, Aug 2026 - Present · 3 mos, Bengaluru, • Lead a team of 10 engineers building B2G smart-city products across road infrastructure, waste management, and property trade domains - owning technical direction, sprint delivery, and client-facing outcomes. • Delivered rapid prototypes and PoCs on tight deadlines using AI-assisted development (Claude), converting business/client requirements into working demos that secured 10+ client approvals/contracts. • Partnered directly with business teams and government stakeholders to translate requirements into technical specs; synced with audit teams on security compliance and audit-log management for government-grade systems. • Drove code review standards and code quality improvements across the team, reducing production defects and release friction. • Architected UI layers surfacing ML model predictions on interactive geospatial dashboards (Kepler.gl, Google Maps) for city-scale infrastructure planning.

Senior Full Stack Engineer at Chitragupta Urban Tech, Feb 2026 - Present · 9 mos, Bengaluru, - Architected and delivered scalable, production-grade web applications using Node.js, React.js, MongoDB, and PostgreSQL, containerized via Docker and deployed on GCP - serving 10k+ users at high request volume. - Built automated ETL/data sync pipelines using cron jobs and batch processing to handle simultaneous background operations, ensuring data consistency and reducing manual intervention across services. - Implemented Redis caching and optimized PostgreSQL query performance through proper indexing strategies and query restructuring, reducing API response latency by ~20–30%. - Established CI/CD pipelines using GitHub Actions and GitHub Packages with integrated security checks, enforcing deployment best practices and reducing release friction across the team.

SDE-2 at Anuvrittiksha Edtech, Dec 2024 - Jan 2026 · 1 yr 2 mos, Remote, - HowToAlgo v1.1 – Performance & Scale: Improved performance 40% (v1.0 → v1.1) via lazy loading and optimized DynamoDB GSIs; lowered p95 latency/TTI; added event-driven flows with DynamoDB Streams and ETL via AWS Glue; scaled to thousands of users. - HowToAlgo v1.2 – Integrity, Cost & Security: Implemented ScaNN plagiarism detection using Bedrock Titan embeddings + vector search using FAISS (replaced O(N²) with top-k); used Gemini Flash for solution clustering, anomaly detection, and auto-summaries; reduced DynamoDB cost 30% by auditing access patterns and right-sizing/adding GSIs; strengthened SEO by improving keywords (+40% ranking boost) & implemented robots.txt + sitemap.xml for better crawlability. - Leadership: Mentored 5 developers; led Agile sprints & code reviews, improving delivery quality and consistency.

SDE-1 at Anuvrittiksha Edtech, Jun 2024 - Nov 2024 · 6 mos, WFH, HowToAlgo v1.0 – Launch: Led design & launch; frontend Next.js/TypeScript/Tailwind, backend Node.js/Express/REST; AWS stack (API Gateway, Lambda, Cognito, DynamoDB); CI/CD with GitHub Actions; Judge0 + secure payment gateways; CloudWatch monitoring; resolved 100% of Critical/High defects pre-go-live. fixed 100% SAST findings and improved security with CSP and stricter headers.

Full-stack Developer at Omninos Solutions, Oct 2023 - May 2024 · 8 mos, Mohali district, Punjab, India, - Live Streaming App – Built a live-streaming platform (React.js, Node.js, MongoDB) using Agora for low-latency video, admin/agency panels, and OTP via Fast2SMS; applied solid system design; shipped via CI/CD to Hostinger/Ultahost. - Chat App – Developed a real-time chat app with Socket.IO/WebSockets and Firebase Realtime Database: 1:1 & group chats, presence/typing indicators, delivery/read receipts, and secure auth.

React Training at Solitaire Infosys Inc, Jul 2022 - Aug 2022 · 2 mos, Mohali

Education

Master's Degree, Artificial Intelligence and Machine Learning, BITS Pilani Work Integrated Learning Programmes

Bachelor of Technology - BTech, Information Technology, Chandigarh Group of Colleges, Landran, Punjab

Public links

github: https://github.com/DIV297

divanshbajaj.netlify.com: https://divanshbajaj.netlify.com/