Cave Digital Australia
Website:
cave.digital
Company:
https://www.linkedin.com/company/cave-digital-australia
Seniority: Mid-Senior level
Industries: Advertising Services
Job details:
About Us:
We are an AI-powered social media management platform built to compete with Hootsuite, Buffer, and Later at a fraction of the price. Creators, marketing teams, and agencies use our product to schedule content, analyse performance, manage social inboxes, run approval workflows, and generate AI captions and images — all from one dashboard. The platform is already built and in active use. It is integrated with major social platforms and it schedules and publishes real posts for real customers today. We are keeping our name off the public posting for now. Full details on the company, product, and traction are shared at first interview.
About This Role — Read This Part Carefully:
This is not a greenfield build. You are inheriting a working product that a five-person team, all with under three years of experience, built without senior technical oversight. They shipped something real under those constraints, which is genuinely impressive — and it also means the platform carries the kind of gaps you'd expect: uneven test coverage, thin observability, security and scale decisions that were never reviewed by anyone more experienced.
Your first job is to find out exactly where those gaps are and fix the ones that matter, while the team keeps shipping features. Your second job is to make those five developers materially better engineers.
If you want a clean slate to architect from scratch, this isn't it. If you're the kind of engineer who enjoys opening an unfamiliar codebase, forming a fast and accurate picture of what's load-bearing versus what's merely ugly, and turning a working prototype into a production-grade system — this is a very good role.
Role Details:
Title: Technical Lead — Full Stack Developer (Web + Mobile + AI)
Experience: 5–7 years, including demonstrated experience leading or mentoring engineers. A formal "Lead" title is not required — informal lead work on a small team counts.
Location: Remote (India). Team and customers span multiple time zones; some overlap expected.
Reports To: Founder & Director
Team: 5 developers today (all under 3 years' experience), growing as the platform scales
Compensation: Based on experience and demonstrated depth
Education: B.Tech / M.Tech in CSE, IT, AI, or ML — or equivalent practical experience
Joining: Immediate joiners preferred
What You'll Own:The First 30–60 Days: Audit and Roadmap:
Assess the existing platform for scalability gaps, technical debt, and security risk. Produce a prioritised remediation roadmap — and then execute it, in sequence, without stopping feature delivery. We expect you to distinguish clearly between what is genuinely dangerous and what is simply inelegant.
The Scheduling Engine:
This is the core of the product and the thing that must never fail. You'll own the background job infrastructure built on Redis and BullMQ: queue design, retry and backoff logic, idempotency guarantees, dead-letter handling, and monitoring. When a social API goes down for an hour mid-morning, the posts queued behind it must be neither lost nor published twice. You'll also own timezone-correct scheduling, including DST transitions, across US, EU, and APAC customers.
Social Platform Integrations:
You own every integration on the platform, end to end:
- Facebook (Pages, Groups)
- Instagram (Business, Reels, Stories)
- Threads
- X / Twitter (API v2)
- LinkedIn (personal, company pages, media uploads)
- YouTube (Data API v3)
- TikTok (TikTok for Developers)
- Pinterest
- Telegram
- Google My Business
That means living with rate limits, quota exhaustion, breaking API changes, deprecations, media-upload quirks, and webhook reliability as a permanent condition of the job — across ten platforms that each fail differently.
Security and Multi-Tenancy:
We hold our customers' social credentials. Multi-tenant OAuth architecture, secrets management, and token rotation are requirements, not nice-to-haves. You'll also be responsible for data-handling practices that stand up to EU and US customer scrutiny.
Web and Mobile Development:
Continue to architect, build, and improve the React web application and the React Native (Expo) mobile app — scheduling, analytics, inbox, collaboration, and reporting modules. Own performance, responsiveness, and accessibility, and handle platform-specific requirements across iOS and Android.
AI Features:
Extend caption generation, AI image creation, hashtag suggestions, best-time-to-post recommendations, and inbox AI replies. Longer term, build autonomous agent workflows that act on a user's behalf. This includes the unglamorous parts: constraining LLM output to a customer's brand voice, evaluating whether that's actually working rather than eyeballing samples, per-user credit metering, and keeping token costs under control at scale.
Production Readiness:
Own the testing strategy, CI/CD pipelines, monitoring and alerting, and incident response for a system processing thousands of scheduled actions per day. You should find out that jobs are failing before a customer tells you.
The Team:
Mentor and level up five junior engineers through pairing and structured code review. Establish PR standards, onboarding, and documentation practices including architecture decision records — phased in without grinding delivery to a halt. The measure of success here is that they get faster and better, not that you personally touch every line. Becoming the bottleneck is a failure mode we will talk about in the interview.
Required Skills:
Core stack:
- Strong JavaScript and TypeScript — clean, typed, maintainable code
- React.js — state management, performance optimisation, component architecture
- Node.js / Express — API design, middleware, error handling
- React Native (Expo a major plus)
- Redis and BullMQ — essential; the scheduling engine depends on it
- PostgreSQL, MongoDB, or MySQL — schema design, indexing, query optimisation
- Advanced REST API design and consumption, including retry, backoff, and rate-limit handling
- Git (GitHub / GitLab)
Production and security:
- CI/CD ownership — GitHub Actions, GitLab CI, or equivalent — configured by you, not just used
- Testing strategy for queue consumers and webhook handlers (unit + integration)
- Monitoring, alerting, and logging — Sentry, Datadog, or equivalent
- Multi-tenant OAuth, secrets management, and token rotation for social logins
- Timezone-safe scheduling logic; awareness of i18n and GDPR-type data-residency constraints
- Engineering process for a growing team — code review standards, onboarding, and documentation practices such as architecture decision records
AI engineering:
- Hands-on work with LLM APIs beyond basic calls — OpenAI / GPT, Anthropic Claude, Google Gemini or equivalent — including prompt design, structured outputs, token management, and cost control
- Judgment about model selection: knowing which model to route a given task to, and why, on cost and quality grounds
- AI-driven product features shipped to real users: content generation, recommendation systems, or autonomous agents
- LangChain or equivalent for multi-step workflows — highly preferred
- Experience constraining LLM output to a brand voice or style guide, with some evaluation loop behind it
Strong plus:
- Deep prior experience with Meta Graph API, LinkedIn API, or X API v2
- Cloud infrastructure (AWS / GCP / Azure) and Docker
- Multi-region hosting or CDN work for latency-sensitive international traffic
- AI-assisted development tooling (Cursor, Claude Code) to accelerate delivery
- Startup experience in a fast-paced, resource-constrained environment
Who Fits Here:
- Ownership-driven — You think past your ticket, you care about the product, and you act like a co-builder rather than a task-taker.
- A clear communicator — You can explain a technical risk to a non-technical founder in a way that helps him prioritise, rather than just sounding alarming — and you give honest status updates.
- Comfortable in ambiguity — Early-stage means shifting priorities and incomplete information.
- Diplomatic about inherited work — You can tell a junior developer their code has a problem without making them defensive about it.
- Product-minded — The goal is user value, not code completion.
- Disciplined asynchronously — Remote-first, accountable without supervision.
What We Offer
- Competitive salary in the band listed above
- Fully remote — work from anywhere in India
- Direct access to the Founder; your technical decisions get made and acted on, not committee-reviewed
- Real ownership of a live product with paying users — not internal tooling or maintenance work
- A small team with no bureaucracy, and the chance to shape both the platform and the engineering org as it grows
Hiring Process
- Founder screening — ownership, judgment, and experience fit. A conversation, not a quiz.
- Technical round — scenario-based architecture discussion covering queue design and failure handling, inheriting an existing codebase, AI workflow design, and leading a junior team. Tied to systems you've actually built, not abstract trivia.
We review applications on a rolling basis and respond to strong candidates within 5 business days.
How to Apply
Email hr@dijura.com with:
- Your resume.
- The completed technical task — https://docs.google.com/document/d/1BKIOI2QbKylnA_ITmMSwzGcBgG7jWiAIkqCDS33mP9Q/edit?usp=sharing.
It should take roughly 2-3 hours; there is no single right answer and we're reading for reasoning, not polish.
- A short note (3–5 sentences) on the most complex background job, queue, or scheduling system you have personally built — real scale numbers, real failure modes, and how you handled retries and idempotency in production.
- If relevant: a link to something you've shipped.
Click on Apply to know more.