Manas Mishra

Backend & AI Engineer · SDE-2 · Founding Engineer

Bangalore, India sdemanas23@gmail.com linkedin.com/in/manas--mishra

Open to backend, AI and senior software engineering roles in Bangalore, Noida and across India.

Summary

Backend and AI engineer (SDE-2) with 5 years of experience (4.5 years full-time plus a 6-month internship) building scalable, cloud-native backend systems, data pipelines and LLM-powered AI agents. Founding engineer at a 0→1 data-analytics startup, owning architecture, deployment and post-launch optimization, and previously technical co-founder of a CA-firm automation startup. Strong in Python, Django, FastAPI, microservices, PostgreSQL, Redis, AWS, Azure and GCP, with hands-on experience in the OpenAI Agents SDK, LangGraph and agentic AI workflows. Focused on performance, reliability and business impact.

Skills

Languages
Python, JavaScript, SQL, Bash
Backend
Django, FastAPI, Flask, Express.js, REST APIs, GraphQL, Microservices, Async workers
AI & LLMs
OpenAI Agents SDK, LangGraph, LLM integration, Agentic AI workflows, Azure Document Intelligence, Machine Learning
Data
PostgreSQL, MySQL, MongoDB, Redis, SQLite, Airflow, ETL / data pipelines, Scrapy, Playwright, Web scraping, Appsmith
Cloud & DevOps
AWS (S3, SQS, Lambda), Microsoft Azure (VMs, PostgreSQL Flexible Server), GCP (App Engine), Docker, nginx, Jenkins, CI/CD, GitHub, Blue-green deployments

Experience

SDE-2 (Founding Engineer), Datavio

Sep 2023 – Present

Bangalore, India

  • Lead backend development for data-analysis services in a 0→1 startup, working directly with the CTO on architecture, deployment and post-launch optimization.
  • Scaled backend infrastructure to handle 100× more load across the database and service layers.
  • Migrated Azure PostgreSQL Flexible Server across regions via read-replica promotion, unlocking 2× server capacity unavailable in the original US region.
  • Built an AI data-analysis agent from scratch on the OpenAI Agents SDK (FastAPI, Redis, Azure) with multi-step reasoning, and agentic AI workflows that automate supply-chain operations.
  • Built ~40 e-commerce and quick-commerce platform integrations, including Playwright connectors for vendor portals with no public API.
  • Built a penny-level payment-reconciliation engine (FastAPI, PostgreSQL) tracing every UTR to invoices, debit notes, credit notes and TDS; 88/88 payments reconciled on a first production run, 99.5% of 1,012 matched on another channel, 0 mismatches on a 982-row validation file.
  • Designed data pipelines on AWS S3, SQS and Lambda that run microservices in parallel, making processing 2× faster.
  • Optimized cron jobs for 10× faster execution, accelerating client onboarding.
  • Introduced versioned blue-green deployments with Docker Compose and nginx: health-gated releases, graceful traffic switching, 3-hour drains for in-flight workflows and one-step rollback.
  • Built a document-AI pipeline (Azure Document Intelligence, rules first with a GPT fallback) that splits debit notes to invoice SKU lines in resumable phases: 94.9% of 486 split cleanly and GPT calls fell from 486 to 2.
  • Designed and deployed microservices in Python, Django and Express.js; optimized PostgreSQL queries, self-hosted MongoDB, scaled Jenkins with Azure VM agents and moved long-running endpoints to async workers.

Technical Co-founder, Bentofi

Apr 2023 – Aug 2023

Startup · CA-firm workflow automation

  • Co-founded a startup automating the manual workflows of Chartered Accountant (CA) firms; owned product engineering and built the MVP end to end.
  • Built Snap & Upload and email-integration intake so receipts and bills flow directly into the books, with an approval window before posting.
  • Automated GST returns and reconciliation, client document collection and bookkeeping data entry; worked with 2 CA-firm clients on the MVP.

Python Developer, APlus Topper

Sep 2021 – Mar 2023

Hyderabad, India

  • Built scalable applications with Django (Python) and SvelteKit, improving performance and reliability.
  • Developed REST APIs enabling efficient data flow and faster feature delivery.
  • Deployed on GCP App Engine (Standard) with CI/CD, reducing release time and increasing uptime.

Projects

llm-graph-optimizer

Python · LangGraph · SQLite

Instruments LangGraph agents in one line, shadow-replays production inputs through a lighter model and reports cost, latency and agreement for each node. Pilot: 91% lower cost ($4.20 → $0.38/month) and 3.2× faster p50 (310 → 96 ms) on a classify node at 98.7% agreement.

SME Agent

Python · LangGraph · SQLite

Conversational agent that turns question papers into question-bank entries, with human-in-the-loop review, Google Drive, Excel and question-bank API tools, and resumable threads checkpointed in SQLite.

Market Intelligence Platform

Django · Scrapy · PostgreSQL · MongoDB · Azure

Tracks trends across marketplaces with ML-driven insights; scraping at scale with proxy management, browser fingerprinting and CAPTCHA handling.

Alerts & Logging Service

Django · Slack SDK · Azure

Real-time error tracking with Slack and email alerts and a centralized logging dashboard for system health.

Psychometric Assessment Platform

Flask · SvelteKit · Google Cloud

AI-driven platform that evaluates candidates' psychometric and cognitive abilities.

Education

B.Tech, Computer Science and Engineering

2019 – 2023

United College of Engineering and Research, Prayagraj · CGPA 8.98/10

International Training Programme, Asian Institute of Technology (AIT)

1 month

Thailand · Selected for a month-long international training programme

Achievements