Deepankar Mahapatro

DevOps Engineer

Bengaluru, Karnataka, India12 yrs 6 mos experience

Highly Stable

Key Highlights

12 years of experience in AI engineering and distributed systems.
Led development of key open-source frameworks like Jina.
Expert in building scalable, cloud-native AI solutions.

Stackforce AI infers this person is a Backend-heavy AI Engineer with expertise in MLOps and cloud-native solutions.

Contact

Skills

Core Skills

Software DesignLarge Language Models (llm)Technical Leadership

Other Skills

Amazon Web Services (AWS)Cloud ComputingComputer ScienceDeep LearningGithub ActionsGo (Programming Language)KubernetesPython

About

Principal Engineer with 12 years of experience in AI engineering, backend, and distributed systems. Led the development of key open-source frameworks like Jina and langchain-serve. Strong expertise in building scalable, cloud-native AI solutions and managing cross-functional teams.

Experience

Goodnotes

Engineering

Oct 2024 – Present · 1 yr 5 mos · Bengaluru, Karnataka, India · Remote

Building production scale AI Agent Backend for millions of users

Software DesignPythonGo (Programming Language)Large Language Models (LLM)

Synthflow ai

Principal Engineer

Jun 2024 – Oct 2024 · 4 mos · Bengaluru, Karnataka, India · Remote

Built a low latency, highly resilient, streaming graph-based orchestration service for Voice AI Agents.
Leveraged LLMs for tool calls, enabling interactions without hallucinations for agents.
Golang | Python | Livekit | LLMs

Amazon Web Services (AWS)Computer ScienceSoftware DesignPythonGo (Programming Language)Technical Leadership+1

Jina ai

Principal Engineer

Sep 2020 – Jun 2024 · 3 yrs 9 mos · Bengaluru, Karnataka, India

Core contributor and maintainer of Jina (20K+ ★), an open-source MLOps framework, and architected Jina Cloud, its SaaS/self-hosting platform.
Implemented Jina’s microservice architecture using gRPC asyncio streaming and protobufs for efficient inter-service communication.
Developed a multi-protocol gateway (REST, WebSockets, gRPC) as the entry point for MLOps pipelines.
Created Kubernetes operators using Golang for cross-cloud deployments and self-hosting.
Built a RESTful API layer and credit tracking system for Jina Cloud using Golang.
Reduced model cold-start times from 5 mins to 30 secs, improving scaling efficiency in Jina Cloud.
Created langchain-serve (1.6K+ ★) to bring langchain apps to production, with OpenAI-based bot integration within Slack, built-in monitoring to track TTFT, latency, cost, and human-in-the-loop features for real-time streaming conversation.
Extensive experience with AWS services including EKS, Batch, Lambda, and Sagemaker, creating scalable, resilient infrastructures for production environments.
Python | Golang | Distributed Systems | Kubernetes | AWS | Serverless | ML/LLMOps | Gen AI

Computer Science