Akilesh Anandraj & Ruben Savio · Professional profileDownload PDF
PROFESSIONAL PROFILE
Akilesh Anandraj & Ruben Savio
Product builders · Technology leaders · AI and automation for institutions, businesses and founders
Akilesh AnandrajFounder & CEO, Greyfeathers Studios Data platforms at Samsung scale · Google · McKinsey
Ruben SavioCTO, Greyfeathers Studios Former Co-founder & CTO, Baaz · R&D, James Hardie (Chicago)
Combined track record100+ products built · 20+ AI applications · software for 23 institutions
Photograph Akilesh Anandraj
Photograph Ruben Savio
PHONE to add
WEB greyfeathers.io
Akilesh Anandraj & Ruben Savio | Professional profile
Executive profile
Builders who turn real operations into working software
Akilesh Anandraj and Ruben Savio are product builders and technology leaders who work as a pair. They run Greyfeathers Studios, an AI-native product studio in Coimbatore, India. Between them they have designed, built and run more than 100 products: consumer platforms at national scale, data systems at Samsung scale, institution-wide administration for a 23-institution education group, fleet operations for hundreds of buses, and AI that reads documents, listens to calls and watches road video.
Their work sits where business operations meet software. They start from how an organisation actually runs (its registers, approvals, vehicles, people and paperwork) and build systems that make that work visible, accountable and faster, with AI used where it removes manual effort and rules kept where decisions must be auditable.
Software is only valuable when the people who run the operation use it every day.
100+Products built across both founders' careers
20+AI applications in production or proof of concept
23Institutions running on systems we built
~300Buses run on our fleet platform
~20,000Students moved by that fleet every day
1MViewers served on a single live stream
$12MAnnual GMV on a commerce platform we built
80+Code repositories behind our AI review gate
13Surfaces we ship on, from web and mobile to car HUDs
Working lens
Understand the operation before writing code. Build the smallest system that changes how people work, put it in their hands quickly, then grow it from what they actually use.
Akilesh Anandraj & Ruben Savio | Professional profile
How we help
What we can do for your organisation
We work with four kinds of partners. In every engagement one of us stays personally involved from the first conversation to delivery, and our studio team in Coimbatore does the building.
Akilesh Anandraj & Ruben Savio | Professional profile
Founder profile
Akilesh Anandraj: data at scale, products from first principles
Photograph Akilesh Anandraj
Founder & CEO, Greyfeathers Studios
Akilesh founded Greyfeathers Studios and leads it today. His background is in large-scale data and engineering: at Samsung he built data platforms handling data at trillion-byte scale, and he has worked with Google and McKinsey.
Among his notable work is a location-triangulation system that infers position from movement signals and uses those patterns to model households and audiences for advertising, the kind of problem where data engineering, algorithms and product judgement meet.
Expertise
Large-scale data platforms and pipelines
Location intelligence and audience modelling
Backend architecture for multi-tenant products
AI systems: model gateways, agents, retrieval
Fintech engines, including a home-loan management engine
Leadership
Founded the studio and grew it on its own products first
Leads client relationships and enterprise engagements
Hands-on in core builds: Godfather, Switchboard, Braid, Klarytee
Sets the studio's AI-native way of building
Keeps the team small, senior and accountable
Products he started before client work
Product
What it does
Year
The Beautiful Menu
Lets a restaurant's menu do the talking: digital menus and ordering
2022
ProfilePot
Profiles collected and kept current, with a hiring-side view
2023
YABA
Personal finance: accounts, loans, expenses, savings and goals
2023
AI Podcaster
An AI podcast host that records and interviews you; the studio's first AI product
Ruben is a mechanical engineer turned product builder. After a master's at the University of Texas at Arlington, where his thesis was filed for patent by CommScope, he spent two years in R&D at James Hardie in Chicago.
In 2018 he returned to India as the first employee of a startup and built its products in sequence: a geofenced ad network in Uber cabs, a hyperlocal video Q&A app, and Baaz, India's first live video shopping platform, which he co-founded and led as CTO.
Stage
Milestone
Education
B.E. Mechanical Engineering, Sri Krishna College, Coimbatore; M.S. Mechanical Engineering, University of Texas at Arlington. Thesis on telecom antenna frame structures, filed for patent by CommScope.
R&D
Two years at James Hardie, Chicago: competitor benchmarking, pultrusion die optimisation and thermal simulation.
2018
Geofenced ad network on tablets in Uber cabs, built with Uber India. Profitable in its first year.
Loka
Hyperlocal video Q&A app: 10,000 users in its first two months.
Baaz
Co-founder and CTO of India's first live video shopping platform: about $12M annual GMV, live streams reaching up to 1M viewers, and an engineering team grown from the first hire to about 100.
Automotive
Software connecting head-up displays to cars, and outdoor-activity apps for smart glasses.
Today
CTO of Greyfeathers Studios: AI-native delivery, institutional platforms, fleet systems and AI products.
In his words
"I'm not Batman and never wanted to be. I'm the best Robin you'll find": strategy, empathy with founders and teams, and getting things shipped.
Akilesh Anandraj & Ruben Savio | Professional profile
The studio we run
Greyfeathers Studios: AI-native, and kept small on purpose
Greyfeathers Studios began as Akilesh's own product studio, building and launching its own products before taking on client work. Ruben joined as CTO and brought a decade of building and scaling consumer and mobility products. Today the studio is a ten-person team in Coimbatore, kept small by choice, with clients in India, the United States, Canada and Singapore.
The team builds with AI at every step of delivery, which lets a small group ship complete products quickly: first versions in days to weeks, not quarters, and production systems that institutions and companies run every day.
What we build
Full-stack products: web, Android, iOS, desktop and backend
Machine learning: vision, language, retrieval and voice
Digitise only after the process is simple. Make status and responsibility visible to the person who decides. Enter information once and reuse it everywhere.
Akilesh Anandraj & Ruben Savio | Professional profile
Education & institutions
A digital backbone for a 23-institution education group
For Vivekanandha Educational Institutions in Tamil Nadu, Greyfeathers builds the systems that run the group day to day: learning, records, correspondence, transport, health services and alumni, one platform across 23 institutions on web, Android and iOS.
Systems in use
Learning management: built around the institution's real structure, from chairman to classroom
Thapal (e-office): inward and outward correspondence, routing, escalation and SLAs
DocuRepo: records with classification, OCR search and versions
Fee tracking, live at several institutions
Live analytics: feature usage per college, for leadership
Operations and campus life
Fleet management for ~300 buses and ~20,000 students a day
Clinic app for the group's healthcare services
Alumni platform connecting graduates to the institution
Canteen operations: menu, kitchen queue, wastage and analytics
File tracking through the people who must touch each file
23Institutions on one platform
Web · Android · iOSRole-based access for every user
One recordEntered once, reused across systems
Automation is valuable only when it simplifies responsibility, improves visibility and strengthens accountability.
Approach
Each module starts from the institution's own registers and approval paths. Differences between colleges are handled as configuration, so every institution runs one common process.
Akilesh Anandraj & Ruben Savio | Professional profile
Fleet operations
Running ~300 buses for ~20,000 students, every day
One system replaced paper registers, route PDFs and phone calls for everyone who keeps the fleet moving: managers, drivers, cleaners, bus in-charges, the workshop, the fuel station and security.
Fleet operations: a morning run with a breakdown, standby dispatch and fuel alerts (from our interactive demo)
Operations
Live GPS on existing trackers; actual vs scheduled time at every stop
Daily route planning: routes merged or split without changing the master route
Breakdowns: nearest standby bus and a valid-licence driver dispatched
Student and in-charge attendance with reconciliation
Maintenance and compliance
Workshop rotation, job cards and maintenance logs; engines and parts traced across buses
In-house fuel station: every fill against the odometer, mileage alerts
Insurance, permits, fitness and licences, renewals flagged 30 days ahead
Gate passes, incident tracking and cost per vehicle
Akilesh Anandraj & Ruben Savio | Professional profile
Automotive
In the car: an AI owner's manual and head-up displays
AI owner's manual for the Volkswagen Taigun and Polo GT (from our interactive demo)
Proof of concept · Volkswagen partner, India
AI owner's-manual assistant
A mobile app with an AI engine over the manuals for the Volkswagen Taigun and Polo GT, every variant and model year. Owners ask in German, English or Hindi and get answers only from the manual of the car they bought, with the page images and steps. A photographed warning light is identified from that model's manual.
Answers cited to the manual page, in three languages.
Automotive · head-up display
HUD software connected to the car
Warnings, road signs and trip information projected in front of the driver. Fully configurable by the client: what is shown, in which font, and where on the display.
Akilesh Anandraj & Ruben Savio | Professional profile
Mobility & location
Location-aware products for cabs, drivers and roads
Geofenced ads in Uber cabsPayana voice co-pilot
Ad tech · with Uber India
Geofenced ads inside Uber cabs
A tablet in each cab connected to an ad platform with the cab's live location. Advertisers drew virtual geofences and bid for them; the ad played when a cab entered a fence, and overlapping fences went to the highest bid. Advertisers included textile brands, cafés, malls, hotels and events.
In-car AI · voice
Payana, a voice co-pilot
Plans a multi-stop drive from a spoken request, compares routes and well-reviewed stops, and hands navigation to Google Maps. It re-checks the route every five minutes against live, historical and forecast traffic, and re-routes without the driver asking.
Payana Engine: dashcam video into GPS-tagged road events
Payana Engine ingests dashcam and IP-camera video, segments it into GPS-tagged events (vehicles, people, congestion, potholes, cracks and speed bumps) and produces an annotated video, a per-second report and a dashboard.
Akilesh Anandraj & Ruben Savio | Professional profile
Wearables, AR & vision
Software for smart glasses, try-on and the camera
Outdoor-activity app on smart glasses: speed, pace, distance, elevation and a live map (from our interactive demo)
Wearables
Smart-glasses activity app
Tracks walking, jogging and running live, with speed, duration, distance, elevation, total climb and a route map shown on the glasses. A sports watch, moved into the wearer's line of sight.
AR & computer vision
Virtual try-on and 3D in the browser
See a garment or a new hairstyle on your own photo before you buy, with AI costs visible before each try. Immersive 3D, AR and VR delivered from a single web link, and a talking 3D avatar loaded from one script tag.
Vision AI
Face indexing and emotion detection
A production-shaped face indexing service on AWS Rekognition, and emotion detection that reads both the face and the voice.
Health data
Wearable health data on Android
Reads health data from Android without integrating every wearable vendor separately, so one app works with many devices.
Akilesh Anandraj & Ruben Savio | Professional profile
AI & automation
AI that reads, listens and checks, with people deciding
Across industries the pattern is the same: AI turns messy input (scanned documents, phone calls, drawings, transcripts, photos) into structured data, and the decisions stay with people and rules that can be audited.
Document AI · POC for SLB
Welding-spec reviewer
Reads scanned, bilingual welding procedure specifications into about 60 parameters and checks each against 26 encoded DNV rules, citing the clause. Aim: a one-day engineer review becomes a one-to-two-hour check-and-sign.
Voice AI
After-hours voice receptionist
Answers calls in natural speech, works out what the caller wants, checks the real calendar and books the appointment. Unusual calls go to a person with the transcript attached.
Vision AI · construction
Drawings to quotation
Reads construction drawings into quantities, then prices them with deterministic rate tables into an auditable bill of quantities: from about 16 hours of take-off to under one.
Agents · hiring
HR agent
Turns résumés and GitHub profiles into scored evaluations, generates assessment tasks and reviews submitted code.
Automation · meetings
Transcript to task, automatically
A meeting bot that joins like a polite attendee, extracts action items from the transcript and files them as tasks.
Knowledge · retrieval
Searchable knowledge from documents
Question answering over policy documents, and a knowledge base built from photographs of whiteboards and notes.
Data quality
Rule-based data validation
Every row of an upload tested against dozens of rules, with clear reasons for each rejection.
Sales · agents
Lead-generation agents
Two AI agents research prospects across the web and score them on buying signals, with a test mode that checks the output before spending.
Akilesh Anandraj & Ruben Savio | Professional profile
Consumer platforms at scale
Products that millions could reach, and businesses could run on
Before the studio, the founders built and ran consumer products where scale, payments and live operations were daily realities. That experience shapes how Greyfeathers builds for clients today.
$12MAnnual GMV at Baaz
1MViewers on a single live stream
~100Engineering team built from the first hire
10,000Users in two months on Loka
Year oneProfit for the in-cab ad network
B2B + B2CMerchant tools and consumer apps
Baaz · live video shopping
India's first live video shopping platform
Merchant app, seller dashboard and marketplace
Payments in production, search and recommendations
Live streams to up to 1M viewers
On-the-ground onboarding with small B2B merchants
Commerce and service products
The Beautiful Menu: restaurant menus and ordering, down to the thermal printer
Two-sided service marketplace: five surfaces on one backend
Multi-vendor travel booking, each side with its own portal
Telesales CRM for outbound teams
HomePe: home-loan management built for borrowers
What it taught us
Scale is an operations problem as much as a technical one: onboarding, support, payments and reliability decide whether a product survives its users.
Akilesh Anandraj & Ruben Savio | Professional profile
Engineering & AI advisory
Helping technology teams ship faster with AI
Beyond building, the founders work with existing engineering teams: diagnosing where delivery slows down, fixing the process, and bringing AI into everyday development in a way the team keeps using.
Engineering review · trading software company
From 20-minute fixes to same-day releases
Fixes took 10 to 20 minutes to write and up to four hours to reach production. The review produced 14 findings and nine changes the team could make in one week: branch protection, a hotfix path, pull-request templates, automatic checks and CI builds.
Recovery plan · consumer platform
Six apps, about 600 open bugs
A structured recovery plan of 41 steps across six apps, ordered so the most painful issues for users were fixed first.
AI in delivery
One standard across 80+ repositories
An AI reviewer checks every pull request against one engineering standard that anyone can edit, with no deploy needed to change it.
AI-native pipeline
Design to working software in days
Design straight to screens, frontend and backend built in parallel, AI-written tests, and feedback pinned on the live product that an AI agent fixes.
What we review
Where work waits between writing code and reaching users
Branching, review, testing and release practices
Who owns what across products and customers
Tooling and cloud spend
What teams get
A short list of changes ranked by impact and effort
Finance & HRLoans, budgeting, hiring and HR platforms
Engagement
What it looks like
Build a product
A first working version in weeks, then continuous improvement with real users
Digitise an institution
Map the registers and approvals, then replace them module by module on one platform
AI and automation
Proofs of concept that become production systems: documents, voice, vision, agents
Work packages
Complete packages (apps, automation, AI, backend) delivered alongside a client's own team
Engineering advisory
A short review of how a team ships, the fixes, and AI set up in its daily work
Working across borders
Clients in India, the United States, Canada and Singapore. Products shipped in English, German and Hindi, with the team working on the client's hours where needed.
We would rather build one system people use every day than ten they admire once.
