Full Stack Engineer

Founding Full-Stack AI Engineer / Platform Lead
Company: Adzzat
Location: Mumbai, Maharashtra / India
Job Type: Full-time
About Adzzat
Adzzat works with AI companies and frontier labs to understand where advanced AI models and agents fail, how those failures can be measured, and how evaluations, benchmarks, datasets, and infrastructure can be built around them.
We are building technical infrastructure for AI evaluation, model failure analysis, benchmarks, human escalation, and enterprise AI evaluation.
About the Role
We are looking for a Founding Full-Stack AI Engineer / Platform Lead who can operate as a founding engineer, platform owner, and technical product leader.
You will work directly with the founders, researchers, and customers to turn AI research and product ideas into production-ready systems.
This is a high-ownership, hands-on role. You will write code, design architecture, build products from scratch, debug complex systems, and help shape Adzzat's engineering direction.
Key Responsibilities
Build infrastructure for large-scale model and agent evaluations.
Develop systems to capture model outputs, agent trajectories, and failure patterns.
Build and manage AI benchmarks, evaluation datasets, rubrics, graders, and evaluation pipelines.
Run evaluations across multiple AI models and agents and compare performance, cost, quality, and reliability.
Build human-in-the-loop and human escalation workflows.
Develop interfaces for human review, expert intervention, approval, correction, and adjudication.
Build model and agent infrastructure including model routing, inference orchestration, fallback logic, experimentation, and reliability monitoring.
Develop customer-facing AI evaluation products for enterprise workflows.
Work with frontier AI labs and research teams to turn identified model failures into scalable evaluation systems.
Build internal AI-native tools and workflows using modern AI coding agents.
Make architectural decisions and establish engineering standards as the engineering team grows.
Participate in technical hiring, code reviews, mentoring, and technical roadmap development.
Required Technical Experience Backend
Experience with several of the following:
Python
TypeScript / Node.js
FastAPI
PostgreSQL
Redis
APIs
Job queues and background workers
Distributed systems
Event-driven architecture
Frontend
React
Next.js
TypeScript
Modern UI frameworks
Analytics dashboards
Data-heavy applications
AI / ML Infrastructure
Strong familiarity with:
LLM APIs
AI agents and tool use
Model evaluation and benchmarking
Agent trajectories
LLM observability
Inference gateways
Model routing
Structured outputs
LLM-as-a-judge
Human-in-the-loop systems
Evaluation harnesses
Cloud & Infrastructure
AWS and/or GCP
Docker
CI/CD
Cloud deployment
Monitoring and logging
Databases
Security fundamentals
AI-Native Engineering
You should be highly comfortable using modern AI coding tools such as Cursor, Claude Code, Codex, GitHub Copilot, and AI coding agents.
We value engineers who can move rapidly from:
Idea → Prototype → Customer Feedback → Production
However, you should also have strong engineering judgment and be able to review AI-generated code, identify architectural problems, debug difficult issues, design scalable systems, maintain security and reliability, and refactor prototypes into production-quality software.
Who We're Looking For
You are likely a strong fit if you:
Have built products from 0 to 1.
Have experience working at an early-stage startup.
Can independently own large technical systems.
Are strong across both backend and frontend development.
Understand AI products and modern AI systems deeply.
Use AI coding tools to significantly accelerate development.
Move quickly and are comfortable with ambiguity.
Have strong product instincts.
Care about good architecture without over-engineering.
Enjoy working directly with founders, customers, and researchers.
Prefer ownership over narrowly defined tickets.
Strong Plus
Experience as a founding engineer.
Built your own startup.
Shipped an AI product used by real customers.
Built AI infrastructure or evaluation infrastructure.
Built AI agent systems or model gateways.
Experience with inference or model routing.
Experience with benchmark infrastructure.
Open-source AI contributions.
Led a small engineering team.
Worked directly with founders or AI researchers.
What Success Looks Like
You will help build a platform that allows Adzzat to:
Identify important AI model and agent failure modes.
Create evaluations around those failures.
Run multiple models and agents against those evaluations.
Validate failures with human experts.
Convert failures into reusable benchmark cases.
Determine when AI systems should escalate to humans.
Continuously evaluate new model versions.
Feed evaluation insights back into customer AI systems.
Why Join Adzzat?
This is an opportunity to become one of the early technical leaders at a company working on the infrastructure required to make increasingly autonomous AI systems more reliable, measurable, and safe to deploy in the real world.
You will have significant ownership and the opportunity to work directly with founders, researchers, frontier AI companies, and enterprise customers.
If you are someone who hears "We think there is an important model failure here" and thinks "Let me figure out how we measure it, build it, and turn it into a product," we'd like to hear from you.

Work arrangement
No