Director, Platform Engineering — Mobius Networks

Distributed systems.
Platforms teams can build on.

12+ years leading cross-functional engineering teams — up to 14 engineers across 4 squads — building cloud-native platforms, distributed systems, and enterprise AI infrastructure. Built and scaled a 12-member Platform Engineering team from the ground up.

View projects Download resume
CKAD Certified badge CKAD Google Cloud Generative AI Leader badge Gen AI Leader SnapLogic Developer Certification badge SnapLogic
12+ Years experience
14 Engineers led · 4 squads
150+ Microservices owned
99.5% Platform availability
Scroll

About

Leading distributed platform and cloud-native engineering teams for over a decade.

Engineering Manager and Platform Architect with 12+ years leading cross-functional engineering teams and building large-scale distributed platforms and cloud-native systems. Experienced leading organizations of up to 14 engineers across 4 Agile squads, partnering with Product, Architecture, SRE, Operations, and Customer Success to turn business objectives into scalable platform capabilities.

Built and scaled a 12-member Platform Engineering team from the ground up, mentoring engineers, growing technical leaders, and supporting promotions into lead roles. My work spans an enterprise DevSecOps automation platform, a sub-millisecond Cloud Data Stream handling 1M+ msg/sec, a distributed DaaS platform across 150+ microservices at 99.5% availability, a unified enterprise AI gateway serving 50+ foundation models, and confidential computing infrastructure for regulated workloads.

Engineering & People Leadership

Built & scaled a 12-member platform team from scratch · Led up to 14 engineers across 4 squads · Mentored engineers into lead-role promotions

Architecture & System Design

Distributed Systems · Event-Driven Architecture · Microservices · API Design · System Reliability · Workflow Orchestration

Proven Delivery at Scale

150+ microservices · 99.5% platform availability · 3x AI delivery throughput · Zero security incidents

2020–Present

Engineering Manager / Platform Architect

Mobius Networks, Gaian Solutions

2017–2020

Software Engineer

Planon Software

2016–2017

Software Engineer

Tech Mahindra

2014–2016

Technical Consultant

IFS Solutions

Professional experience

Building distributed platforms and leading engineering teams at scale.

Engineering Manager / Platform Architect

Mobius Networks, Gaian Solutions Apr 2020 – Present

Own architecture, engineering roadmap, and team leadership for multi-tenant distributed platforms and AI/ML infrastructure in regulated enterprise environments, leading cross-functional squads through delivery, quarterly planning, and production readiness.

Team Building & Engineering Leadership

Built and scaled a 12-member Platform Engineering team from the ground up across Platform, DevOps, and Infrastructure disciplines. Led 3 Engineering Leads and 14 engineers across 4 Agile squads on RunRun, driving quarterly planning, architecture reviews, and incident management. Mentored engineers into promotions to lead roles.

12-member team built 14 engineers / 4 squads Engineers promoted to lead

Distributed Platform Architecture

Designed multi-tenant distributed platform with event-driven communication patterns enabling asynchronous processing, improved resilience, and container-based runtime on Kubernetes across 150+ microservices and 12+ databases.

150+ microservices Event-driven 99.5% availability

Enterprise AI Gateway & Generative AI Platform

Led a team of 6 engineers delivering a production-grade Enterprise AI Gateway supporting 50+ foundation models, improving GPU utilization 30–50% and cutting inference costs 25–40%. Led a further team of 4 engineers on a Generative AI platform (Image/Video/3D Studio) serving 1000+ enterprise marketplaces on NVIDIA H100 clusters, cutting infra costs 35–40% and lifting throughput 3x.

6-engineer team · 50+ models 4-engineer team · 1000+ tenants 3x throughput

Security & Compliance

Implemented GDPR/PII governance, mTLS, IAM compliance programmes, and confidential compute (CVM/TEE) infrastructure achieving full audit readiness for regulated clients.

GDPR compliant mTLS everywhere Zero incidents

Software Engineer

Planon Software Dec 2017 – Apr 2020

Designed enterprise integration services connecting cloud platforms, enterprise applications, and operational systems for facility and asset management.

Enterprise Integration Services

Designed microservices and real-time integration APIs enabling cross-system asset and facility data synchronisation between Azure, SAP, and enterprise platforms.

Microservices Real-time sync Multi-cloud

Software Engineer

Tech Mahindra Jun 2016 – Dec 2017

Delivered Java-based REST services and integration components for enterprise platforms, with focus on performance optimisation.

Java REST Services & Performance

Built Java REST services and integration components for enterprise platforms. Optimised SQL queries and execution plans, improving enterprise reporting performance by 40%.

Java / REST 40% faster reports SQL optimisation

Technical Consultant

IFS Solutions Jul 2014 – Jun 2016

Developed ERP integrations using Java, SQL, and service APIs across enterprise systems.

ERP Integrations

Developed and maintained ERP integration components using Java, SQL, and service APIs for enterprise business systems across industries.

Java / SQL Service APIs ERP systems

Agentic AI & GenAI platform engineering

Leading engineering teams building production LLM and agentic infrastructure at enterprise scale.

Enterprise AI Gateway

Team of 6 Engineers 50+ Foundation Models

Architected and led engineering delivery of a production-grade Enterprise AI Gateway. Defined platform architecture, technical roadmap, and governance standards for intelligent model routing, session memory, multi-agent orchestration, batching, caching, and fine-tuning.

Inference Stack & Model Serving

Served self-hosted models (gpt-oss-20b, gpt-oss-120b, Llama 3.1) on NVIDIA H100 via vLLM, Ollama, and LoRAX, with ONNX models served through Triton Inference Server. Autoscaled GPU pools to zero with KEDA between traffic bursts.

vLLM · Ollama · LoRAX Triton (ONNX) H100 + KEDA scale-to-0

Token Cost & Caching Strategy

Applied Anthropic/OpenAI-style prompt batching and prompt caching to cut redundant token spend across repeated context. Used Cache-Augmented Generation (preloading static context directly into the model's KV cache) as a lower-latency alternative to Retrieval-Augmented Generation for latency-sensitive request paths.

Prompt caching CAG vs. RAG 25–40% lower inference cost

Impact

Improved GPU utilization by 30–50%, reduced inference costs by 25–40%, and enabled rapid onboarding of enterprise AI applications while ensuring scalability, security, and production reliability.

+30–50% GPU utilization -25–40% inference cost

Generative AI Platform

Team of 4 Engineers 1000+ Enterprise Marketplaces

Led architecture, platform strategy, and engineering execution for an enterprise Generative AI platform powering Image Studio, Video Studio, and 3D Studio, delivered across 1000+ enterprise marketplaces on NVIDIA H100 Kubernetes clusters.

Engineering Standards & Delivery

Established engineering standards, sprint planning, design reviews, and production readiness processes while improving engineering velocity across the team.

Design reviews Production readiness gates

GPU Scheduling & Cost

Optimized GPU utilization through intelligent scheduling and scale-to-zero capabilities, reducing infrastructure costs by 35–40%, increasing GPU utilization by 40%, and improving throughput by 3x.

-35–40% infra cost +40% GPU utilization 3x throughput

Featured projects

Enterprise-grade platforms, shipped and running in production.

Distributed Database-as-a-Service Platform

Multi-tenant DaaS integrating 12+ database technologies across transactional, analytical, time-series, geospatial, and streaming workloads, processing 10TB+ of data daily.

Apache Flink PostgreSQL MongoDB TiDB Apache Hudi
10TB+ Daily 500K–1M Events/sec 12+ DB Technologies
View Details

Enterprise LLM Inference Gateway

Led a 6-engineer team delivering a unified gateway for multi-provider LLM and on-premise AI inference, supporting 50+ foundation models with intelligent routing and GPU infrastructure management.

Anthropic OpenAI vLLM KServe GPU MIG
6-Engineer Team 50+ Models +30–50% GPU Utilisation
View Details

RunRun - DevSecOps Platform

Owned architecture and delivery for a cloud-native self-service infrastructure provisioning platform. Led 3 Engineering Leads and 14 engineers across 4 Agile squads using Camunda + Terraform.

Camunda Terraform Vault ArgoCD
14 Engineers / 4 Squads Zero-downtime Multi-cloud
View Details

ML Platform Engineering (Nesy Factory)

Owned strategy and delivery for an enterprise ML Platform with a 15-person cross-functional org (3 Engineering Leads, 6 ML Engineers, 3 Backend, 2 UI, 1 UX), building reusable MLOps/LLMOps capabilities from data pipelines to serving.

Kubeflow MLflow KServe PyTorch
15-Person Org 3x Delivery Speed 15+ Prod ML Workloads
View Details

Cloud Data Stream (CDS)

Ultra-low-latency microservice communication platform using RSocket TCP

RSocket Kafka Flink Kubernetes
<1ms Latency 1M+ Msg/sec Real-time
View Details

CVM Confidential Infrastructure

Confidential VM and Trusted Execution Environment (TEE) platform for regulated enterprise workloads

AMD SEV-SNP Intel TDX Vault Istio mTLS
Zero Incidents Full Audit Ready GDPR Compliant
View Details

Upcoming Projects

AI platform ideas in active development

LLM Gateway

Unified wrapper around self-hosted and hosted LLM backends behind one API.

Ollama vLLM LoRAX ONNX Anthropic OpenAI Gemini
In Development

Image Studio

Wrapper unifying image-generation models behind a single interface.

Qwen Stable Diffusion FLUX
In Development

Video Studio

Text-to-video generation wrapping the WAN model family, with multiple model support (TI2V-5B, T2V-A14B) and customizable output resolution. Ships as a lightweight Docker image that downloads models on first use, exposed through a RESTful API with GPU-accelerated inference.

WAN TI2V-5B T2V-A14B
Text-to-Video GPU-Accelerated RESTful API

3D Model Generator

Wrapper around the Hunyuan3D model for text/image-to-3D-asset generation.

Hunyuan3D
In Development

Technical expertise

12+ years across platform architecture, AI infrastructure, and engineering leadership.

Team Leadership
Built & scaled a 12-member platform team · Led 14 engineers across 4 squads
System Architecture
Distributed systems · Event-driven design · Microservices
Cloud & Platform
Kubernetes (CKAD) · AWS/GCP · Terraform · GitOps
Workflow Orchestration
Camunda BPM · Distributed workflows · Saga patterns
AI/ML Infrastructure
vLLM, Ollama, LoRAX, Kubeflow, MLflow · Google Cloud Generative AI Leader
Security & Compliance
mTLS, IAM, GDPR · Confidential computing (CVM/TEE)
Data & Streaming
Kafka, Flink, PostgreSQL, MongoDB, TiDB

Career targeting

Role match analysis, open to remote and global opportunities.

Senior Engineering Manager, Platform / AI
97%
Strongest match · Target role · Immediate fit
Director of Platform Engineering
96%
Current role · Team leadership + architecture
Engineering Manager, AI/ML Platform
94%
High demand · Google Cloud Generative AI Leader certified
Principal / Platform Architect
93%
Strong fit · Deep architecture focus
VP of Platform Engineering
87%
Natural progression · Broader scope
CTO – Growth-stage Startup
78%
Ambitious stretch · High upside

Personalized learning paths, message drafts, and job tracking.

Personal journey

Beyond code — learning, exploring, and staying grounded.

Courses & Certifications

Google Cloud Generative AI Leader

Google Cloud

Certification demonstrating expertise in generative AI strategy, implementation, and leadership on Google Cloud infrastructure.

Issued Feb 2026 · Expires Feb 2029
Verify on Credly
CKAD Certification

Certified Kubernetes Application Developer (CKAD)

The Linux Foundation / CNCF

Advanced Kubernetes application development, deployment strategies, and cloud-native patterns.

Certified Dec 2024
Verify on Credly
SnapLogic Developer Certification

SnapLogic Developer Certification

SnapLogic

Integration platform development: pipelines, connectors, and enterprise data flow design on SnapLogic's iPaaS.

Certified
Verify on Credly

Leadership Skills & Pre-MBA Statistics

IIM Ahmedabad

Executive leadership development and quantitative foundations, investing in the people-management side of the Engineering Manager track.

Ongoing
Learning How to Learn

Learning How to Learn

Coursera, University of California San Diego

Mental tools to master tough subjects and develop effective learning techniques applicable to complex engineering domains.

Completed

Books That Shaped My Thinking

Outliers Book

Outliers

Malcolm Gladwell

"Success is not just about individual merit but opportunity, timing, and cultural background."

The Alchemist Book

The Alchemist

Paulo Coelho

"When you want something, all the universe conspires in helping you to achieve it."

A Man Called Ove Book

A Man Called Ove

Fredrik Backman

"All people at root are time optimists. We always think there's enough time to do things with other people."

Tech Evangelism & Innovation

ML Model Grok4 Advocacy

Promoting the adoption of advanced AI models in enterprise environments. Exploring the potential of Grok4 for real-world applications, from natural language processing to automated decision-making systems.

AI/ML LLMs Enterprise AI

Technology Critique & Analysis

Critical analysis of emerging technologies, focusing on practical implementation challenges, scalability concerns, and real-world impact. Advocating for responsible technology adoption.

Tech Analysis Innovation Best Practices

Travel Adventures

🇮🇳 India Explorations

Pushkar, Rajasthan

Sacred lakes, colorful markets, and spiritual vibes

Kufri & Shimla

Mountain serenity and colonial charm

Hampi, Karnataka

Ancient ruins and architectural marvels

Goa

Beaches, culture and vibrant nights

Rishikesh

Yoga capital with serene Ganges views

Nanded

Spiritual heritage and history

Hyderabad

City of pearls and modern tech

Philosophy & Mindfulness

Art of Living

Practicing mindfulness and breathing techniques to maintain balance in the fast-paced tech world. The Art of Living philosophy teaches us that happiness is our nature, and stress is just a temporary state that can be managed through proper breathing and meditation.

"When you are fully in the present moment, you drop all your worries and anxieties."

Core Values I Live By

Compassion in Code
Work-Life Harmony
Continuous Growth
Community Building

Fitness & Wellness Journey

The Effects of Running

Running isn't just physical exercise, it's mental therapy. It clears the mind, improves focus, and provides the mental resilience needed for complex problem-solving in software architecture.

Mental Clarity

Enhanced focus for debugging complex systems

Stress Management

Natural stress relief from high-pressure deadlines

Creative Problem Solving

Many breakthrough solutions come during runs

Consistency Over Intensity

Just like in software development, consistency beats sporadic bursts of activity. A regular running routine has taught me the value of small, daily improvements that compound over time, a principle I apply to both code and career growth.

5+ Years running
3-4x Weekly runs
10K+ Best distance

Open source projects

Personal projects and utilities on GitHub.

Cordova RFID Connector

Cordova plugin for Zebra RFD8500 and TSL RFID scanners.

Cordova Zebra RFD8500 TSL
View on GitHub

Forge Viewer App

Web app embedding Autodesk Forge Viewer for 3D/BIM model visualization.

Autodesk Forge
View on GitHub

Career Insight App

Backend service powering career/job-search analysis and insights.

Node.js
View on GitHub

RTMP Server

Nginx-based RTMP server for live video streaming.

Nginx RTMP
View on GitHub

RSocket

Low-latency socket connections for streaming data and messaging.

RSocket
View on GitHub

Socket.io Message Rooms

Real-time message rooms backed by Redis Pub/Sub and Kafka.

Socket.io Redis Pub/Sub Kafka
View on GitHub

Get in touch

Open to Director / Engineering Manager, Platform & AI roles.

Ready to build something amazing?

I'm always interested in challenging projects and opportunities to work with innovative teams.

Email

gurpreetgandhi3@gmail.com

LinkedIn

linkedin.com/in/gandhigurpreet

GitHub

github.com/grupret

Career Assistant

Engineering Manager Journey · AI-Powered

GitHub

LeetCode

Today's Tasks & Goals

Daily Brief

  Optional: enable Power Mode for deeper, open-ended Q&A

Quick questions work instantly with no key. Get a key at console.anthropic.com for freeform mode. Stored locally only.

Today's Coding Challenge

Exercism Practice Tracks

IQ Challenge: Systems Thinking

+10 pts

EQ Exercise: Leadership Scenario

Communication Mastery

Focus Timer: Pomodoro

25:00

Ready to focus, pick a session above

  Quick Notes