Skip to content
the forward pass0%

Global Top 10

ICC NIUM Hackathon

National Finalist

Smart India Hackathon

Published Researcher

IJANA / ASICS 2024

AI Systems, End to End

Across product, infra, and ML

[ 02 / REPRESENTATION ] · what the network learned

Designed for recruiters, research leads, and product teams evaluating real execution depth.

Comfortable moving between AI system design, architecture, experimentation, and delivery.

Best work sits where applied AI, usable products, and fast iteration meet.

layer 01

AI Workflow System Design

I think in workflows, evaluation loops, and user decisions, not just model calls. That matters when an AI product has to feel reliable, legible, and genuinely useful.

layer 02

Full-Stack Delivery

I move from product idea to working system across frontend, backend, APIs, orchestration, testing, and deployment. The value is speed without handing off the hard parts.

layer 03

Research-Grounded Execution

I can operate in build mode and investigation mode: comparing approaches, pressure-testing decisions, and translating what works into something usable for teams, judges, and end users.

[ 03 / INFERENCE · selected work ]

The DeploymentLog

Trained in theory. Deployed in Dubai — production systems with real stakes.

[ 03 / INFERENCE ] · live systems · real stakes

The Deployment Log

browse with arrows, drag, or swipe — the page scrolls on regardlessthe log grows

[ 04 / TRAINING ] · how the weights formed

Proof of Work

Tested under competition, research, and institutional scrutiny — every epoch lowered the loss.

0+

Hackathons

0

Countries Built From

0

Publication

Top 0

Global Finish

International

Global Top 10

ICC NIUM Hackathon

Web3 & Blockchain

National

National Finalist

Smart India Hackathon 2023

Government Problem Statement

Regional

Winner

Let's Hack India Tour

Parul University

Regional

Top 8 + Incubation Offer

Vadodara Police Hackathon

Incubation by PIERC

Government

Certificate of Appreciation

Vadodara Rural Police

From IPS Rohan Anand for AI prototype demo

checkpoint saved · peer-reviewed

AI Cloud Service Recommendation System (Eva)

IJANA Journal, ASICS Conference · 2024

Co-authored ML-based cloud service recommendation model achieving 70%+ accuracy on custom 1,000+ data point dataset across 6 user-defined parameters.

[ 05 / LATENT SPACE ] · related skills live near each other

Capability Map

Capabilities first, tools second. Six clusters of the space — every tool here has shipped real work.

Languages
06
PythonJavaScriptTypeScriptC/C++SQLHTML/CSS
AI & Machine Learning
12
PyTorchTensorFlowscikit-learnOpenCVMediaPipeNumPyPandasSciPyReinforcement LearningLSTM & AttentionDeep SVDDExplainable AI (XAI)
LLM & Generative AI
05
Google Gemini APILLM Inference PipelinesPrompt EngineeringResponsible AIAI Chatbot Development
Web & DevOps
09
ReactViteFastAPISQLAlchemyDockerGitGitHub ActionsREST APIsJupyter Notebooks
Developer Tools
04
CursorGitHub CopilotClaude (Anthropic)Git/GitHub
Domains
07
Real-Time SystemsEdge DeploymentMLOpsBiometric AuthenticationWeb3 & BlockchainSmart City SystemsData Analytics

six clusters · one connected system

[ 06 / BACKPROP ] · the past updating the present

Where the Weights Formed

Academic grounding for applied AI systems — currently mid-training in Dubai.

  • Masters in AI with Business

    SP Jain School of Global Management

    Dubai, UAE · 2025 - 2027

    Current

    GPA: 3.6 / 4.0

    Machine LearningDeep LearningNeural NetworksNatural Language ProcessingOptimizationOperations Management
  • B.Tech in Computer Science & Engineering (AI)

    Parul University

    Vadodara, India · 2020 - 2024

    GPA: 8.04 / 10

    University Representative, National Entrepreneurship Challenge, IIT Bombay

[ 07 / ROUTING ] · where the signal goes next

Ways To Work Together

Three routes through the decision boundary — hiring, research, and product work.

For Recruitersroute 01

Applied AI roles with delivery depth

Built for teams hiring AI engineers and builders who can turn requirements into working systems with strong product execution.

For Research Collaborationroute 02

Product-facing applied research

Strong fit for conversations around intelligent systems, multimodal interfaces, and applied experiments that can move beyond papers.

For Product Buildsroute 03

AI tools and product MVPs

Available for select collaborations involving internal AI workflows, developer-facing tooling, and fast prototype-to-product iterations.

[ 08 / OUTPUT ] · the next model is always in training

Got a project, an idea, or a role worth building for?

Let's talk.

accepting new signals · AI roles, research, product builds

kartik@forward-pass:~$ new-conversation

avg_response_time < 24h