Founding Engineer (Backend / AI)

TaskHarmony; India; On-site; INR 1,200,000 - 1,800,000; Full time

Benefits offered by this job

Open‑source style collaboration

Opportunity to work with LLMs in prod

Job summary

TaskHarmony seeks an experienced backend engineer to build core systems from zero and ship scalable, multi‑tenant APIs. You will work on unstructured email understanding, LLM‑driven data extraction, and end‑to‑end schema API monitoring in a fast‑paced startup environment.

You’ll own production‑grade pipelines with async Python, Postgres, and cloud deployments on Azure, while collaborating across teams to deliver reliable, scalable software at speed.

Qualifications

• Have at least 2 years of real, full‑time backend experience (fresh graduates or candidates without full‑time work experience should not apply).
• Have shipped and supported production systems.
• Work independently without constant hand‑holding.
• Want to work on LLMs in production, not demos or tutorials.
• Prefer ownership, responsibility, and speed over comfort.

Responsibilities

• You’ll be building core systems from zero.
• Systems that understand unstructured business emails.
• LLM‑powered extraction of freight rates and shipment data.
• Prompt versioning, accuracy tracking, and observability.
• Gmail / Outlook / IMAP integrations at scale.
• Event detection, video clipping, and evidence generation.
• Cloud edge orchestration with retries and fault tolerance.
• Multi‑tenant APIs used by enterprise customers.
• Async Python services with PostgreSQL.
• CI/CD pipelines with fast, confident production releases.
• You’ll own a feature‑rich end‑to‑end schema API deployment monitoring.
• You won’t be maintaining legacy code.

Skills

Python

FastAPI

Async PostgreSQL

SQLAlchemy

Docker

GitHub Actions

Azure

React

TypeScript

LLMs in production

Tools

OpenAI

Claude

Langfuse

YOLOv8

OpenCV

FFmpeg

Job description

Job Overview

• You’ll be building core systems from zero.
• Systems that understand unstructured business emails.
• LLM‑powered extraction of freight rates and shipment data.
• Prompt versioning, accuracy tracking, and observability.
• Gmail / Outlook / IMAP integrations at scale.
• Event detection, video clipping, and evidence generation.
• Cloud edge orchestration with retries and fault tolerance.
• Multi‑tenant APIs used by enterprise customers.
• Async Python services with PostgreSQL.
• CI/CD pipelines with fast, confident production releases.
• You’ll own a feature‑rich end‑to‑end schema API deployment monitoring.

Responsibilities

• You won’t be maintaining legacy code.
• You’ll be building core systems from zero.

Agentic Email Intelligence (Freight Forwarding)

• Systems that understand unstructured business emails.
• Multi‑agent pipelines: fetch, classify, extract, auto‑respond.
• LLM‑powered extraction of freight rates and shipment data.
• Prompt versioning, accuracy tracking, and observability.
• Gmail / Outlook / IMAP integrations at scale.

Warehouse Video Intelligence

• Distributed pipelines processing CCTV feeds.
• Event detection, video clipping, and evidence generation.
• Cloud edge orchestration with retries and fault tolerance.

Scalable Backend Infrastructure

• Multi‑tenant APIs used by enterprise customers.
• Async Python services with PostgreSQL.
• Cloud deployment on Azure (containers, queues, storage).
• CI/CD pipelines with fast, confident production releases.
• You’ll own a feature‑rich end‑to‑end schema API deployment monitoring.

Tech Stack (Your Daily Weapons)

• Backend: Python 3.11+, FastAPI, async PostgreSQL, SQLAlchemy.
• AI / LLMs: OpenAI, Claude, agentic workflows, Langfuse.
• Computer Vision: YOLOv8, OpenCV, FFmpeg.
• Cloud: Azure (Container Apps, Functions, Service Bus, Blob Storage).
• DevOps: Docker, GitHub Actions.
• Frontend (working knowledge): React, TypeScript.

Requirements

• Have at least 2 years of real, full‑time backend experience (fresh graduates or candidates without full‑time work experience should not apply).
• Have shipped and supported production systems.
• Work independently without constant hand‑holding.
• Want to work on LLMs in production, not demos or tutorials.
• Prefer ownership, responsibility, and speed over comfort.

What Makes You Stand Out

• Built async systems, queues, workers, or event‑driven backends.
• Taken an LLM feature from prototype to real usage.
• Comfortable using AI coding tools (Cursor, Claude Code).
• Startup experience or high‑ownership roles.
• You say you built this, not your team built it.