%20--%3e%3ccircle%20cx='32'%20cy='32'%20r='21.5'%20fill='none'%20stroke='%23F4F4F5'%20stroke-width='1.2'%20stroke-dasharray='2.8%204.2'%20stroke-linecap='round'%20opacity='0.42'/%3e%3cline%20x1='39.8'%20y1='27.5'%20x2='46.7'%20y2='23.5'%20stroke='%23F4F4F5'%20stroke-width='4'%20stroke-linecap='round'/%3e%3cline%20x1='39.8'%20y1='36.5'%20x2='47.1'%20y2='40.8'%20stroke='%23F4F4F5'%20stroke-width='4'%20stroke-linecap='round'/%3e%3cline%20x1='27.5'%20y1='39.8'%20x2='23.3'%20y2='47.2'%20stroke='%23F4F4F5'%20stroke-width='4'%20stroke-linecap='round'/%3e%3cline%20x1='24.2'%20y1='27.5'%20x2='16.4'%20y2='23'%20stroke='%23F4F4F5'%20stroke-width='4'%20stroke-linecap='round'/%3e%3ccircle%20cx='32'%20cy='32'%20r='9'%20fill='%23F4F4F5'/%3e%3ccircle%20cx='51'%20cy='21'%20r='5.2'%20fill='%23F4F4F5'/%3e%3ccircle%20cx='51'%20cy='43'%20r='4.6'%20fill='%23F4F4F5'/%3e%3ccircle%20cx='21'%20cy='51'%20r='4.6'%20fill='%23F4F4F5'/%3e%3ccircle%20cx='13'%20cy='21'%20r='4'%20fill='%23F4F4F5'/%3e%3c/svg%3e)
FAQ · 72 answers · No gates
Questions about building software with BalochDev
Direct answers on services, AI, pricing, timelines, stack, and how we start — every answer on this page in full. Senior team. AI in the toolchain. Code you own.
2024Founded
20+Specialists
14+Core team
AIIn the toolchain
About BalochDev
Who we are, where we’re based, and how we work.
BalochDev is an AI-first custom software studio. We design and build web apps, mobile apps, and AI products for international clients — and we advance Balochi language technology with community partners. Claude Code, agents, and modern frameworks sit inside our toolchain so a focused senior team ships faster without skipping review.
We are a remote-first studio rooted in Balochistan, working with clients across the Middle East, Europe, North America, and beyond. Optional meetups in Qatar, Dubai, Bahrain, and the region when face-to-face helps.
BalochDev has been shipping since 2024 — client products, AI systems, and Balochi language technology in parallel. The studio is young by design: modern stack, AI in the toolchain, and a clear dual mission from day one.
We work with 20+ senior specialists across engineering, design, content, and ops. Our core delivery team is 14+ people — the same faces in kickoff as in standups months later. No junior bench shuffled onto your project.
Claude Code, agents, and modern frameworks sit inside our toolchain on every project. They compress scoping, build, and review loops — they do not replace senior architecture, compliance judgment, or code ownership.
Both. We ship MVPs for founders and production features for operators inside larger companies. The scoping conversation is the same: clear problem, honest fit, fixed milestones when scope is stable.
No. Most engagements are commercial web, mobile, and AI products for international clients. Balochi language technology is our highest-priority community mission — funded alongside client work, not a requirement for hiring us.
Senior people on every project, AI in the toolchain (not just the pitch), code in your repo from day one, and honest “no” when we are not the right fit. You get continuity and ownership — not a B-team handoff.
Services & AI
What we build — web, mobile, agents, RAG, and automation.
Custom software across web, mobile, and AI: web applications, cross-platform apps, LLM integrations, RAG, agents, chatbots, UX/UI, MVPs, workflow automation, and production hosting on modern stacks.
LLM integrations, RAG over your documents, AI agents, chatbots, extraction pipelines, recommendations, and intelligent workflow automation — for first AI features and for teams adding AI to mature products.
RAG (Retrieval Augmented Generation) connects large language models to your own data — docs, databases, knowledge bases — so answers are grounded in company-specific information instead of generic web knowledge.
A chatbot answers and routes conversations. An agent can plan steps, call tools (CRM, docs, calendars, APIs), and complete multi-step work with guardrails. Many products need both.
We build custom LLM chatbots that understand context, handle complex queries, and answer from your data — far beyond rigid decision-tree templates. Deploy on your site, app, or messaging channels with consistent context.
Yes — that is a large share of the work. We add search, assistants, generation, or automation without breaking what already works, with staging reviews and guardrails.
Yes — automation that replaces repetitive ops glue (spreadsheets, ad-hoc Zapier chains) with purpose-built flows, often combining AI with your existing tools and APIs.
Dashboards, CRMs, marketplaces, internal tools, customer portals, MVPs, RAG platforms, and AI-enabled products — custom builds on modern infrastructure, not configured templates.
Yes. We start with a quick audit (and analytics when available) before redesigning. Most failed redesigns treat the UI as the problem when the real issue is inconsistent patterns or unclear information architecture.
Yes. From discovery and design through frontend, backend, and deployment — you get a production-ready product, not a deck someone else has to build.
AI-first development
How AI actually shows up in the work — honestly.
Engineers use production-grade tools — Claude Code, Cursor, Copilot — inside the codebase. AI handles boilerplate and first drafts; senior engineers own architecture, business logic, and quality.
No. Tools like Claude Code operate in the repo — not copy-paste from a chat window. Every meaningful change is reviewed by an experienced engineer before it ships.
Typically much faster on the mechanical layer — scaffolding, boilerplate, tests, and first-draft code — so focused MVPs that once took months can land in weeks when scope is clear. Quality gates stay the same.
Every line that ships is reviewed. Most bugs come from ambiguous requirements, not the tool — which is why we invest in a clear spec before accelerating the build.
Yes. Code review, testing, staging, and security-minded checks still apply. AI accelerates writing; humans ensure quality and ownership.
Primarily Claude Code, with Cursor and GitHub Copilot as needed. They accelerate work across React, Next.js, Node, Python, and the rest of our stack.
We pick per project — commonly Claude, OpenAI, Gemini, and others when fit. You keep ownership of accounts and billing where that matters for compliance.
It does not own architecture decisions, compliance docs, senior reviews, or final UX judgment. Engineers own those; AI compresses the loops between them.
No. You keep your repo, product, and org chart. We compress delivery loops with senior people still accountable.
Web & mobile
Websites, web apps, PWAs, and store-ready mobile.
Both. Marketing sites and SEO-aware brochure experiences, plus multi-role web apps with auth, dashboards, payments, and APIs — typically React/Next.js with a production backend.
Web apps live in the browser with logins and business logic. Native/cross-platform apps ship to App Store / Play Store with device APIs. PWAs install to the home screen, can work offline, and skip store review — a fit when store fees or review friction hurt the product.
iOS and Android — usually React Native or Flutter for one codebase, both stores. Native Swift/Kotlin when platform-specific APIs demand it; PWAs when stores are optional.
Depends on your team, timeline, and existing stack. React Native fits JS/TS shops; Flutter shines for highly custom UI consistency. We recommend after discovery, not a slogan.
Yes — listings, screenshots, privacy labels, and submission. First submissions are rarely approved on both stores the same day, so we budget a resubmit cycle into timelines.
Both. Most projects include API, database, auth, and cloud deploy. We do not hand off half an app.
Technical SEO foundations are part of delivery: clean structure, performance awareness, metadata, and crawlable routes. Ongoing content SEO can be added after launch.
Yes. Responsive, mobile-first layouts are standard — every critical flow is checked across phone, tablet, and desktop breakpoints.
Yes. We audit conversion and UX bottlenecks, then modernize for speed, clarity, and mobile performance — with your brand intact.
Process & engagement
Sprints, discovery, timelines, and how builds run.
Fixed-milestone builds, discovery/scoping sprints, embedded work inside your repo, and ongoing retainers when feasible. Most clients start with discovery or a fixed MVP once scope is clear.
Clear milestones with working software you can click — weekly demos, staging before production, and reviews in your repo rather than status decks alone.
Discovery turns a rough idea into a concrete scope — typically days to about two weeks — producing a brief stable enough to estimate and fix-price the build.
Common. We run a short Scope + Spec phase first. If scope stays genuinely ambiguous, we use milestone time-and-materials with transparent logging until it stabilizes.
Yes. We can work in your repo, match conventions, and hand off to your engineers — embedded delivery without locking you into a forever maintenance contract.
Focused MVPs often land in weeks to about 1.5 months from a clear brief. Larger SaaS, commerce, or AI platforms run longer with fixed milestones and weekly demos.
A single AI feature into an existing app can ship in a few weeks. Broader AI products (RAG, agents, multi-channel) commonly run 6–12+ weeks depending on data and integrations.
Cross-platform MVPs typically land in the multi-week to few-month range including store prep. PWAs can be faster when stores are not required. Complexity and integrations drive the range.
A clear owner on our side, async updates (Discord or your channel), and demos on a predictable cadence. Budgets are tracked per milestone, not as open-ended hourly fog.
Pricing & estimates
How we quote — and how to get a fast range.
Loaded, all-in milestone numbers — not opaque rate cards. When scope is locked you get a fixed quote. Use the free AI estimate for a fast ballpark, then we refine on a call or proposal.
Yes. It gives a fast cost range for websites, web apps, mobile, or AI features. It is a starting point — discovery turns it into a concrete quote.
AI accelerates boilerplate and first drafts — the hours that usually inflate agency bills. You still pay for engineering judgment, architecture, and review; fewer wasted hours to the same output.
Typically yes — milestones start after agreement and an initial payment or platform escrow (e.g. Upwork) when you prefer that path.
Clear SOW/NDA, then milestones, deposit, or freelance platforms. International clients are normal — written English delivery and timezone overlap are standard.
Not for locked scope. If you change scope mid-build, we re-estimate the delta before work continues — no silent burn.
You do. Code lives in your repository from the first commit. We retain no rights to client work and do not reuse your code across other clients.
Technology & stack
What we ship with — and what we take over.
React, Next.js, TypeScript, Node, Python/FastAPI, Supabase/Postgres, Cloudflare Workers & Pages, Vercel, and major LLM APIs. Boring, proven tools for the parts that must not fail.
Yes — Supabase for Auth, Postgres, RLS, and realtime; Cloudflare for Workers, Pages, and edge delivery. A common pairing on BalochDev products.
Often yes, after a short audit. We stabilize what is salvageable, then ship milestones — better than a full rewrite when the core is sound.
When they fit a fast MVP or migration path. For long-lived products we usually recommend owned code so you are not locked into a platform’s ceiling.
No. We prefer infrastructure you control — your Cloudflare, Vercel, or cloud accounts — with credentials and runbooks handed over.
Environment isolation, least-privilege keys, and no secrets in the client. Sensitive work gets staging reviews and sensible defaults before production.
Working with us
Communication, handoff, and post-launch support.
Weekly demos, staging URLs, Discord (or your channel), and protected branches. You see real progress every week — not a black-box dump at the end.
Remote-first with overlap for Middle East, Europe, and many US hours. Async updates fill the gaps so progress does not wait on a single standup.
Yes, when scoped — monitoring, dependency updates, small features, and hosting ops. Hand-off to your team is always available; maintenance is optional.
Your repo, environments, credentials you own, deploy notes, and a walkthrough. The goal is that your team (or a future hire) can continue without us.
Yes. We match conventions, leave clear PRs, and teach the workflow so someone internal can continue after we leave.
Warranty windows and retainer support can be scoped per project. Critical production issues get priority when we are on a support agreement.
Balochi mission
Language technology with the community — not over it.
Advancing Balochi language technology — keyboards, browser and office support, ethical AI, and community partnership — while delivering client software that funds the mission.
With educators and native speakers — never over them. Work with partners like Balochi Academy stays grounded in real usage, not extractive data grabs.
Yes. Client delivery has clear milestones; language initiatives run as parallel tracks with their own partners and goals. One does not silently steal budget from the other.
Getting started
The fastest paths from idea to first milestone.
Run the AI estimate, send a proposal, or contact us for a short discovery call. Bring the product idea, constraints, and a rough budget — we will say honestly whether we are the right fit.
The problem you are solving, who the users are, any must-have integrations, timeline pressure, and a budget range. Rough wires or competitor links help — polish is optional.
Yes — discovery, competitive scans, and clickable prototypes so you can test assumptions before committing to a full build.
We will say so early and, when we can, point you toward a better path. A bad engagement costs both sides more than an honest no.
Browse the portfolio for selected products, or start on About to meet the team. Live links are listed when clients allow public demos.