Puneet Gupta | AI Builder | AI Deploy Network

About Puneet

Hands-on engineering leader and co-founder/CTO of EmailZap, an AI-native Gmail app with 2,000+ monthly active users. I design the architecture, write the code and fix the hard production problems.

At EmailZap I built Zap, a LangGraph multi-agent system with tool calling, MCP access and a personalised memory layer, plus a 15-model eval harness, weekly drift detection and inference optimisations that cut LLM cost by 86%. I also shipped PII pseudonymisation and CASA Tier 2 certification.

Before that: Lead Backend Engineer at Certa, where I worked onsite with Dun & Bradstreet in New York on a white-label compliance platform used by 20+ enterprises, and founding engineer at Level AI, building a real-time voice assistant (ASR, NER, RASA).

Nine years across three startups. I care about what we build, what it costs to run and whether customers pay for it.

Professional profile

Location: Mumbai, India
Time zone: Asia/Calcutta
Work mode: Remote, Hybrid, On-site
Availability: Available Now
AI experience: 9+ years
Industry experience: 9+ years
Languages: English, Hindi
Working hours: 08:00-19:00, Mon-Fri
Typical response: 12
Preferred engagement: Medium (multi-team programmes)
Travel: International travel
Available in: India

Proven on the network

Implementation records and publicly available work, where published.

2 AI Deployments delivered, including 2 publicly identified implementations available to inspect.

Reported business outcomes

Hours Automated160 hours
period not specified
Automated 160 hours (period not specified)
Cost SavingsUSD 452,800
period not specified
Reduced operating costs by USD 452,800 (period not specified)
Revenue ImpactUSD 1,000,000
period not specified
Generated USD 1,000,000 additional revenue (period not specified)

Builder-reported outcomes, recorded on deployment and case study records that AI Deploy Network reviewed before publication. AI Deploy Network does not currently capture a measurement basis or calculation methodology for reported outcomes, so none is shown.

What Puneet builds

Implementation domains listed on this profile. Evidence-supported items are identified separately below.

Specialisations

Agentic Workflow Automation

AI Agents

Document Intelligence

Multi-Agent Systems

RAG & Knowledge Retrieval

Builder-listed deployment types

AI Agent
AI Evaluation Framework
AI Orchestration
Code Generation
Code Review Assistant
Internal Developer Tooling
Personalization Engine
Task Automation
Workflow Automation

Show all deployment types

Workflow Automation

AI Deployments

Real AI implementations demonstrated through evidence.

01 / AI DeploymentPublished record

AI-native SDLC for a food-tech company's India engineering pod: AI-assisted development and automated code review

Built and led a two-engineer India pod for a $100M+ revenue food-tech business on an AI-native delivery workflow, with AI-assisted development and automated code review. Ticket-to-PR lead time fell from about three days to one.

160 hours automated / per month · USD 250K cost savings / per year · USD 1M reported revenue impact / one-time

Food & Beverage · Code Generation · Code Review Assistant · Internal Developer Tooling · Workflow Automation

View Deployment

02 / AI DeploymentPublished record

Zap: LangGraph multi-agent email and calendar assistant for EmailZap

Production AI agent inside EmailZap, an AI-native Gmail app. Zap answers questions about a user's inbox and automates Gmail and Google Calendar tasks, backed by a memory layer, PII pseudonymisation and a multi-model eval harness.

USD 202.8K cost savings / per year

Cross-Industry / Horizontal · AI Agent · AI Evaluation Framework · AI Orchestration · Personalization Engine · Task Automation

View Deployment

Implementation capabilities

Evidence-supported means linked to a published Deployment or Case Study; other capabilities are Builder declared.

Evidence-supported

AI Agent2 records
AI Evaluation Framework2 records
AI Orchestration2 records
Code Generation2 records
Code Review Assistant2 records
Internal Developer Tooling2 records
Personalization Engine2 records
Task Automation2 records
Workflow Automation3 records

Show all evidence-supported capabilities

Personalization Engine 2 records
Task Automation 2 records
Workflow Automation 3 records

Builder declared

Agentic Workflow Automation
AI Agents
Document Intelligence
Multi-Agent Systems
RAG & Knowledge Retrieval

Technology

Technologies listed by the Builder, distinguished by published evidence.

Evidence-supported

AWS2 records
Anthropic Claude2 records
FastAPI2 records
Google Cloud2 records
LangGraph2 records
Langfuse2 records
Model Context Protocol2 records
OpenAI GPT2 records
PostgreSQL2 records
Python2 records

Builder declared

Django
Docker
LangChain
Qdrant
Redis
Terraform

Industry experience

Published Deployments support the listed counts. Other industries are Builder declared.

Cross-Industry / Horizontal

1 published Deployment

Food & Beverage

1 published Deployment

IT Services & Systems Integration

Builder declared

Software & SaaS

Builder declared

Case Studies

Published narratives of implementation and results.

Professional Services · Lead Generation · Lead Qualification · SDR Agent · Workflow Automation

AI-in-the-loop outbound pipeline in n8n: lead discovery, qualification and outreach built in two days, run in ten minutes a day

A four-workflow n8n pipeline that finds B2B buyers from public signals, qualifies them with Claude, verifies emails and drafts first lines, with a human approving every send. Built in two days; operated in about ten minutes a day at cents per lead.

Read case study

Food & Beverage · Code Generation · Code Review Assistant · Internal Developer Tooling · Workflow Automation

AI-native SDLC for a food-tech company's India engineering pod: AI-assisted development and automated code review

Built Thistle's India engineering pod on an AI-native SDLC. We first tried a lights-out factory with no human review, which did not hold up. We replaced it with smart review: agents had to prove their changes and split them into core and leaf code, and humans reviewed only the core. Ticket-to-PR lead time fell from about three days to one.

Read case study

Software & SaaS · AI Agent · AI Evaluation Framework · AI Orchestration · Personalization Engine · Task Automation

Zap: LangGraph multi-agent email and calendar assistant for EmailZap

Zap is EmailZap's AI agent for Gmail and Google Calendar. I designed and shipped the first version, grew its usage about 3.7x, and cut inference cost 86%. Production then showed that agent quality depends on the harness, not the model, so I led a contract-first rebuild that makes the rules executable before any model is involved.

Read case study

Trust & verification

A closer look at how this profile and its evidence are reviewed.

Community Member

AI Case Studies verified

Published evidence is reviewed against the AI Deploy Network Evidence Standard.

How verification works

Community

Verified Builder

Independently reviewed by AI Deploy Labs against the platform Evidence Standard.

AI Case Studies Verified

How this is verifiedWhat the platform has established, and what remains builder-declared.

Identity and organisation: Identity checks confirm that the person presenting this profile controls the account and the professional identifiers attached to it, such as a LinkedIn profile. Organisation membership is confirmed by the organisation, not self-declared by the builder.
AI Deployments: An approved AI Deployment is an implementation record the builder submitted and the platform reviewed against the Evidence Standard before publication. Approval confirms the record is complete, coherent and consistent with the supporting material provided — it is not an audit of the client's systems.
Business outcomes: Outcome figures are recorded by the builder on the deployment or case study that produced them, and are reviewed at publication. Aggregates shown on this profile are calculated from those approved records only.
Evidence: Supporting evidence comes from the records the builder published on this platform: approved AI Deployments, published AI Case Studies, verification outcomes and organisation participation.
Currency: Older evidence continues to count and is never removed. Recency is presented separately through Professional Currency so that current expertise can be assessed alongside historical work.
Traceability: Every figure on this profile can be followed to the record that produced it. Open the linked AI Deployment or AI Case Study to read the outcome exactly as it was reported.

The platform does not independently audit client systems or financial records. How this is verified

Professional record and further evidence

Client Portfolio

Portfolio started

Studio Management LLC (EmailZap)
Agentic AI • Software & SaaS
United States2026
Took EmailZap, an AI-native Gmail app, from concept to 2,000+ monthly active users. Built Zap, a LangGraph multi-agent assistant for email and calendar tasks with MCP access and a personalised memory layer, plus an eval harness across 15 models and inference optimisations that cut LLM cost per user by 86%.
1 AI Deployment1 AI Case Study
Cost
Currency not specified
View Related AI Deployments →
Thistle
AI-assisted engineering • Food & Beverage
United States2025
Fractional engineering lead for a $100M+ revenue food-tech business. Set up and led the India engineering pod and introduced AI-assisted development and automated code review, cutting ticket-to-PR lead time from about three days to one while revenue grew 30% with no extra engineering hires.
1 AI Deployment1 AI Case Study
Hours
160
Cost
Currency not specified
Revenue
Currency not specified
View Related AI Deployments →

Professional Certifications

Some recognition

Game Theory
Stanford Online and The University of British Columbia (Coursera)
Verified Credential
Issued Nov 2018
Credential ID: RNYEY4QRWWPL
View Credential
Social and Economic Networks: Models and Analysis
Stanford Online (Coursera)
Verified Credential
Issued Nov 2018
Credential ID: 6AV9EY5RKYQR
View Credential

Ready to discuss your AI implementation?

Tell Puneet Gupta what you are trying to automate and the outcome you need. Your message goes directly to the builder, with a copy kept in your Organisation Workspace.