Brief
Send a brief or hop on a 30-min call. We listen for what you actually need versus what you're asking for.
Hepha Works · A HephaTech studio
Open for new engagements
Hire designers, developers, and AI engineers — all under one studio.
A small India studio shipping brand, web, mobile, AI & computer vision, marketing, tech recruitment, and fractional CTO work. Scoped honestly, shipped on time, billed transparently.
Eight services. One studio.
Click any category to start a brief. We'll come back with a written scope and a fixed number within 1–2 days.
Selected work.
A few of the ships out the door. Three more under NDA aren't shown here.
Brand · Web
Stealth AI consumer app
Identity system and one-page launch site for an AI consumer app entering private beta.
Mobile · AI
Habit tracker · AI coach
Native React habit tracker with offline SQLite, weekly AI summaries, and an archetype identity layer.
Mobile · Marketing
Quiz app for Gen Z
Anonymous-by-design app with a share-card pipeline, chain tracking, and 4,500+ pre-authored packs.
Want yours featured here? Early case studies often get a discount in exchange for a public write-up.
Engagements.
Fifteen scoped gigs. Each is one focused outcome with a fixed timeline and a price quoted on brief.
Brand
Logo Design
3 concepts, 2 rounds of refinement, full delivery in vector + brand-ready exports.
View detailsBrand
Poster Design
Single poster, social-ready and print-ready exports. Great for events and launches.
View detailsBrand
Brand Identity Starter
Logo + type + colour + 20-page brand guidelines. Everything you need to launch.
View detailsWeb
Landing Page
Single-page marketing site, designed and built. Mobile-first, SEO-ready, deploy on us.
View detailsWeb
Multi-page Website
Up to 5 pages, CMS-backed, built for editing without a developer. Deploy included.
View detailsMobile
Mobile App MVP — Design
Up to 8 high-fidelity screens, prototype, and a written handoff for a developer.
View detailsAI
AI Prototype
A working prototype — agent, RAG bot, or AI feature — wired into your data with a clear demo.
View detailsAI
LLM App / Agent Build
A production-ready LLM app or agent — RAG, tools, evals — built on Claude, OpenAI, Gemini, or open models.
View detailsAI
Computer Vision / Industrial AI
Computer-vision pilot for industrial use — defect detection, OCR, inspection, or quality grading.
View detailsData
ML / Data Analytics Sprint
One question answered with data — model, dashboard, or pipeline — with a written report.
View detailsMarketing
SEO Audit
Technical & content SEO audit with a 30-page report and a prioritised fix list.
View detailsConsulting
Codebase Audit
Full-stack codebase & security review with a written report and a 60-min walkthrough.
View detailsHiring
Talent Sourcing Sprint
5 vetted candidates for one role in two weeks. Sourcing, screening, and intro calls.
View detailsConsulting
Business Strategy Call
90-minute deep dive on a single decision — pricing, GTM, hiring, stack — with written notes.
View detailsConsulting
Consult Call
60-min call to scope a project or unblock a decision.
View detailsHow it works.
Four steps from brief to ship. Same shape whether it's a 5-day gig or a 12-week build.
Scope
Within 1–2 days we email a written scope: deliverables, milestones, timeline, and a fixed number for the engagement.
Build
Milestone billing, weekly progress updates, and a working preview link from day one. Change requests get scoped, never absorbed.
Ship
Final review, deploy, handoff docs, and a written summary. Optional standing retainer after launch — no obligation.
Frequently asked questions.
Answers to the questions we get most before a kickoff call.
How long does a typical project take?
Landing pages run 1–3 weeks. Brand systems take 2–4 weeks. Web and mobile apps span 4–12 weeks for a focused first release. AI builds vary widely — a prototype in days, a production system in weeks.
Can we pay in instalments?
Yes. Once we've scoped the engagement, projects typically bill by milestone — 30–40% to start, then per delivered milestone. Sprints split 50/50. Retainers are billed monthly in advance.
Which payment methods do you accept?
India: direct bank transfer, UPI, or PO-based invoicing for organisations. Global: wire transfer or international card. GST-compliant invoices for everything. Every invoice follows a written scope and signed proposal — no payment-link checkout.
What stacks do you typically use?
Web: Next.js, React, plain HTML/CSS when it fits. Mobile: React Native and Flutter. Backend: Node, Python, Postgres, serverless. AI: Claude, OpenAI, Gemini, open-source LLMs, RAG via vector DBs and agent frameworks. Computer vision: PyTorch, OpenCV, YOLO, ONNX for industrial AI like defect detection, OCR, and quality inspection.
Do you build computer vision and industrial AI systems?
Yes. Defect detection, OCR, object counting, quality grading — pilots and production systems. Stack is typically PyTorch, OpenCV, YOLO for vision; Claude, OpenAI, Gemini, open-source models with vector DBs for LLM apps.
Do you handle talent acquisition?
Yes. Talent Sourcing Sprints find and screen 5 vetted candidates for one role in two weeks. Longer-term tech recruitment is available as a retainer for engineering, design, and AI roles.
Are you taking new work right now?
Yes — we're open for new engagements. The status pill at the top of the page reflects current availability.
Do you sign NDAs?
Yes. We sign mutual NDAs before any sensitive brief is shared. Several engagements on this site stay under NDA — we don't show what we can't show.
Start a project.
Tell us what you're building. We'll come back with a written scope and a number within 1–2 days.