Associate: Quality & Release Engineer — AI Product

About Goonj
We are an early-stage, stealth-mode AI product company building a voice-first conversational
platform for young professionals from diverse backgrounds. Our vision is to help users make
more money, elevate their careers, and improve their social standing, all while nurturing
emotional well-being. We want to deliver real-world impact to underserved communities who feel
left behind.
Role Overview
We are hiring a full-time Associate: Quality & Release Engineer - AI Product to be the go-to
person for delivering our product to our users. You will own three things end-to-end: (1) testing
across web and voice surfaces, (2) running CI/CD and deployment, and (3) helping wire the
React/TypeScript client to backend services as features develop.
You will work directly with an external development vendor who handles core engineering and
an external development vendor who handles design and UI/UX. Your job is to make sure their
output reaches our users intact — bugs caught early, builds delivered cleanly, and the front-end
faithfully connected to the back-end.
The stack is current (Python, React/TypeScript, AWS, voice AI), the team is small, and the work
moves fast.
The role is based out of Lucknow.
Key Responsibilities
● Product Testing
Run end-to-end test passes on each release — voice conversations, web client
flows, curriculum behaviour, edge cases around interruption, latency, and audio
quality.
Validate that user progress tracking, state management, and safety systems
behave exactly as specified.
Verify AI-driven features (modules, conversational simulations, content) adapt
correctly to the regional and cultural context our users come from.
Reproduce reported bugs cleanly, file tight reports with repro steps and logs, and
verify vendor fixes before they deliver.
Build out a regression checklist over time and grow it into automated coverage
where it makes sense.
Confirm vendor-delivered features actually match the product spec — not just the
ticket.
● CI/CD and Deployment
Own the build, packaging, and deployment workflow across our services (Python
backends + React client). Keep it boring, repeatable, and fast.
Manage Docker images, environment configs, and deploys to staging and
production on AWS.
Triage build and deploy failures, investigate root cause, and improve the pipeline
so the same failure doesn't happen twice.
Keep '.env'-style configs and secrets clean across environments.
● UI Integration
Help wire the React/TypeScript client to backend APIs (REST and WebSocket)
as features are delivered.
Fix integration bugs on the client side — auth flows, session state, voice events,
error states.
Pair with the vendor team when the boundary between frontend and backend is
the source of an issue.
● Vendor Coordination & Communication
Serve as the primary technical point of contact between the vendor and our
internal team on quality, release, and integration matters.
Translate technical issues and trade-offs into accessible language for
non-technical stakeholders (founder, content team).
Flag risks early — escalate before they become release blockers.
Stay current with GenAI advancements and flag opportunities to strengthen the
product.
Required Skills & Experience
1+ year of professional experience building or delivering software.
Comfortable in **Python**— can read and modify backend code well enough to debug
an integration issue.
Comfortable in **JavaScript/TypeScript and React**— can wire a component to an API,
handle async state, and deliver a small UI fix.
Hands-on with **Git, Docker, and at least one CI tool** (GitHub Actions, GitLab CI,
CircleCI, etc.). If you've never run a deploy yourself, this isn't the right role.
YAML/JSON literacy — our product is configuration-driven; you'll need to read and
validate complex YAML specs to test what the product actually does.
Working understanding of how LLM-based applications fit together end-to-end (prompt
assembly, context management, model calls, safety layers) — enough to test them, not
necessarily build them.
Solid bug-reporting instincts: clear repro steps, expected vs actual, environment, logs.
Honest, direct communication — especially when something is broken or unclear. We
work with an external vendor; clean communication is a daily requirement.
Preferred Skills (Not Required)
Exposure to **LLM APIs** (OpenAI, Gemini, Anthropic) — prompt engineering, API
integration, or conversational AI pipelines.
Basic understanding of voice AI pipelines (ASR/TTS, WebRTC, voice-bot frameworks).
AWS hands-on (DynamoDB, Redis, Lambda, ECS, Cognito) and operational monitoring.
Familiarity with state machines and event-driven architectures.
Experience working with external development vendors or vendor-delivered codebases.
Exposure to regional language processing or culturally adaptive AI products.
A working side project you actually delivered — tutorials don't count.
What Success Looks Like
First 30 days: You can run the full stack locally, you've delivered one production deploy,
and you've filed your first solid bug report against a vendor delivery.
First 90 days: Our release process is documented, our regression checklist is real, and
the UI is no longer where most integration bugs hide.
First 6 months: You're the person on the team everyone trusts to deliver safely. New
features can't go to production without your sign-off, and you've earned that role through
judgment, not gatekeeping.
How We Work
Small team, in-person in Lucknow, direct collaboration with the founder.
Stealth product — you'll see the full architecture and roadmap from day one, but
everything stays internal.
AI-literate culture: use the tools (Claude, Cursor, Gemini, etc.) openly and well. We care
that you understand what you deliver, not whether you typed every character.
We assume good faith, straight talk, and a delivery-focused mindset.

Work arrangement
No