Prateek Bhardwaj profile photo 1 on Tal

Prateek Bhardwaj

Software Engineer @ alltius | Backend & AI | Python · Node.js · LLM Agents · RAG · Distributed Systems

India

4+ years of work experience

alltius.ai
Aug 20251 yr 1 moPresent

Bengaluru

Designed and built a multi-agent system using LangGraph from scratch — including tool-use agents, RAG pipeline, and prompt evaluation harness to validate output quality before shipping to production • Rebuilt the document ingestion pipeline to support large S3 sources using durable workflow orchestration (Temporal) to handle failures, retries, and partial completions in long-running jobs. • Added incremental ingestion and checkpoint-based retry to the S3 pipeline, so a failed mid-run job resumes from its last checkpoint rather than re ingest from scratch, critical for sources with hundreds of large files. • Managed EKS deployments including on-premises deployment for a client, monitoring via ArgoCD and Datadog, and manual scaling during memory-intensive ingestion runs.
Wayground (formerly Quizizz)
Jun 20232 yrs 2 mosJul 2025

Bengaluru

Owned the backend for the Comprehension question type -a nested question structure. Designed the MongoDB schema and built all CRUD APIs. • Led monolithic-to-microservices decoupling initiative: extracted core business logic into standalone Node.js services with API versioning and service-to-service communication patterns, reducing deployment time by 60%. • Shipped LLM-powered features to production (quiz generation, question enhancement, recommendations)
MicroSec
Jan 20225 mosMay 2022

Created Docker-based microservices for Flask and Postgres powered API-based projects. Worked with RabbitMQ to subscribe to sniff and analyze data packets from different network protocols. Engineered and initialized the graph-based database service with Nebula Graph.
SirpiDataScience
Jun 20213 mosAug 2021

Lead the Monthly and Weekly Reports project and worked on deploying the latest changes Developed the Non-UI portion for the Reports and lead the project to be used internally for testing purposes
SirpiDataScience
Oct 20204 mosJan 2021

Engineered web-based shiny apps used by wind engineer clients to generate monthly and weekly wind reports Developed the R Markdown script which forms the basis for the whole project. Worked with several different projects across the board like Testing API and chatbots
GitHub
@prateekb1912

Prateek’s AI Stack

antigravity
antigravity
claudecode
claudecode

Education

Indian institute of information technology una

Bachelor of Technology - BTech, Computer Science

Indian institute of information technology una

20194 yrs2023

Join Prateek and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Prateek Bhardwaj on Tal

Headline

Software Engineer @ alltius | Backend & AI | Python · Node.js · LLM Agents · RAG · Distributed Systems

About

I build things that scale and things that think. High-traffic backend systems - microservices, distributed caching, GPT integrations. Multi-agent AI systems and fault-tolerant pipelines - Langgraph, Temporal, EKS, 10k+ PDFs ingested in a single flow. 3+ years building high-traffic backend systems and multiagent AI pipelines. Reach me at prateekb1912@gmail.com

Location

India

AI stack

antigravity

claudecode

Work experience

Software Engineer II at alltius.ai, Aug 2025 - Present · 1 yr 1 mo, Bengaluru, Designed and built a multi-agent system using LangGraph from scratch — including tool-use agents, RAG pipeline, and prompt evaluation harness to validate output quality before shipping to production • Rebuilt the document ingestion pipeline to support large S3 sources using durable workflow orchestration (Temporal) to handle failures, retries, and partial completions in long-running jobs. • Added incremental ingestion and checkpoint-based retry to the S3 pipeline, so a failed mid-run job resumes from its last checkpoint rather than re ingest from scratch, critical for sources with hundreds of large files. • Managed EKS deployments including on-premises deployment for a client, monitoring via ArgoCD and Datadog, and manual scaling during memory-intensive ingestion runs.

Software Engineer I at Wayground (formerly Quizizz), Jun 2023 - Jul 2025 · 2 yrs 2 mos, Bengaluru, Owned the backend for the Comprehension question type -a nested question structure. Designed the MongoDB schema and built all CRUD APIs. • Led monolithic-to-microservices decoupling initiative: extracted core business logic into standalone Node.js services with API versioning and service-to-service communication patterns, reducing deployment time by 60%. • Shipped LLM-powered features to production (quiz generation, question enhancement, recommendations)

Back End Developer at MicroSec, Jan 2022 - May 2022 · 5 mos, Created Docker-based microservices for Flask and Postgres powered API-based projects. Worked with RabbitMQ to subscribe to sniff and analyze data packets from different network protocols. Engineered and initialized the graph-based database service with Nebula Graph.

Intern at SirpiDataScience, Jun 2021 - Aug 2021 · 3 mos, Lead the Monthly and Weekly Reports project and worked on deploying the latest changes Developed the Non-UI portion for the Reports and lead the project to be used internally for testing purposes

R Programming Intern at SirpiDataScience, Oct 2020 - Jan 2021 · 4 mos, Engineered web-based shiny apps used by wind engineer clients to generate monthly and weekly wind reports Developed the R Markdown script which forms the basis for the whole project. Worked with several different projects across the board like Testing API and chatbots

Education

Bachelor of Technology - BTech, Computer Science, Indian institute of information technology una

Public links

github: https://github.com/prateekb1912