Skip to content
AKSHAY KANKAL

I build AI that enterprises can trust. And hardware that saves energy.

Engineering lead. I build AI platforms that answer honestly and refuse correctly, and power electronics that cut what factories waste.

Ready to relocate

Try the demoEmail me

Try it

Most AI demos show you an answer. This one shows you a refusal.

Pick a department, ask a question, watch which documents are even allowed into the answer.

You are signed in as

Illustrative demo with sample documents. No real company data.

Pipeline

  • FY26 travel and expense policyfinance
  • Supplier payment terms, EUfinance

Submit the claim in the expense portal within 30 days of travel, attach receipts over 25 euro, and your cost centre owner approves it.

How I think

What I believe about building this.

My first paid work was in 2015, a website for the Cyber Cell at the Washim District Police, while I was still studying. I co-founded a software company at university and ran it for four years. I have been building since.

  1. 01

    The idea and the design should come from the engineer. AI should refine and speed up the execution. Engineers keep ownership of every decision and every line that ships.

Employees reachedBrose AI assistant, approximate
20,000
Messages per dayQuestions and answers, approximate
50,000
LocationsIn 25 countries
76
DevelopersOn the AI-SDLC harness, approximate
50

Built

Three systems I designed and built.

Each one had to work for people who never asked for AI, and hold up under access rules written long before it.

01Brose

Enterprise AI assistant

An internal generative AI assistant used by about 20,000 employees across 76 locations in 25 countries.

  • Answers questions from company documents for Finance, HR, Controlling and Shopfloor.
  • It acts as well as answers. Employees raise incidents, security incidents and support tickets through it, and check their status.
  • Permissions are enforced at retrieval through Azure AD groups. An evaluation framework tests answer quality and access correctness, and gates every release.
Read the detail
Architecture diagram for Enterprise AI assistantAzure ADEmployee. Select to see why it is here.EmployeeAssistant. Select to see why it is here.AssistantRetrieval. Select to see why it is here.RetrievalDocuments. Select to see why it is here.DocumentsActions. Select to see why it is here.ActionsIncidents and tickets. Select to see why it is here.Incidents andticketsEvaluation framework gates every release

Select a box to see why it is there

02CreditRiskMonitorIn pilot

MCP server for regulated credit data

Credit scores, payment scores and peer analysis exposed to AI clients as tool calls.

  • A Model Context Protocol server that turns credit risk data into tool calls an AI client can make directly.
  • Authentication, entitlement checks and audit logging on every call.
  • Per customer isolation and rate limits, so one customer never sees another customer data and no single integration can crowd out the rest.
Read the detail
Architecture diagram for MCP server for regulated credit dataTool callAI client. Select to see why it is here.AI clientMCP server. Select to see why it is here.MCP serverAuth and entitlements. Select to see why it is here.Auth andentitlementsScores and peers. Select to see why it is here.Scores and peersIsolation and rate limits. Select to see why it is here.Isolation andrate limitsEvery call is written to an audit log

Select a box to see why it is there

03CreditRiskMonitor

AI-SDLC harness

A controlled environment for AI assisted development, used by about 50 developers.

  • Built on the Model Context Protocol. Every developer works inside the same controlled environment for AI assisted development.
  • Defines which tools, code and data an agent is able to reach.
Read the detail
Architecture diagram for AI-SDLC harnessDeveloper. Select to see why it is here.DeveloperAI-SDLC harness. Select to see why it is here.AI-SDLC harnessAllowed tools. Select to see why it is here.Allowed toolsAllowed code. Select to see why it is here.Allowed codeAllowed data. Select to see why it is here.Allowed dataAgent. Select to see why it is here.AgentThe harness decides the reach, not the agent

Select a box to see why it is there

Built with

  • Python
  • JavaScript
  • Node.js
  • C#
  • .NET
  • React
  • Next.js
  • SQL
  • PostgreSQL
  • Redis
  • Docker
  • Azure
  • Azure OpenAI
  • MCP
  • RAG
  • Agentic workflows

Hardware and IoT

Not everything I build runs in a data centre.Some of it sits in a factory and saves power.

I have worked on energy monitoring since 2016. At CodeElectra we built a wireless electricity consumption monitoring and control system over IoT, which is where the hardware started.

Varnilix Private Limited is a DPIIT recognised industrial IoT company I founded in May 2024. It builds V-GRID, distributed power factor correction hardware that helps factories use energy more efficiently.

DPIIT recognisedIndustrial IoTPower electronics

Published patent applications

01Distributed autonomous reactive power compensationPublished application202621022494
02Power factor correction apparatus with deterministic hardware safety interlocks and relay weld detectionPublished application202621016587

Named inventor. These are published applications, not granted patents.

  • ResearchIoT energy monitoring deviceMAT Journals · 2026
  • RecognitionEmergency Vehicle Navigator, an IoT ideaIntel Higher Education, shortlisted in a national selection, 2017

Timeline

Ten years of shipping, in short.

  1. 2015First paid workWashim District PoliceBuilt the new website for the Cyber Cell at the Superintendent of Police office, while still studying. Recognised with a commendation letter that December.
  2. 2017Co-founder and directorCodeElectra TechnologiesCo-founded a software company at university and ran it to August 2021. Our own product was a wireless electricity consumption monitoring and control system over IoT. Incubated at the Savitribai Phule Pune University incubation centre in 2019.
  3. 2020B.E. Computer EngineeringSavitribai Phule Pune UniversityGraduated First Class from G. H. Raisoni Institute of Engineering and Technology.
  4. 2021Associate, Education, Training and AssessmentInfosysTrained and mentored more than 500 engineers worldwide, and built the online assessment system used across the programme.
  5. 2023Senior Software EngineerBroseIncident reporting and automation for the security operations and IT service management teams across Brose sites. On site in Bamberg, Coburg and Wuerzburg.
  6. 2024Technical LeadBroseLed four engineers building the internal generative AI assistant.
  7. 2024Founder and directorVarnilix Private LimitedFounded a DPIIT recognised industrial IoT company building V-GRID.
  8. 2026Engineering LeadCreditRiskMonitorLeading the India engineering team of 15 at a NASDAQ listed credit risk company, engaged through Encore Infotech Services.

Contact

Let's build something that ships.

Ready to relocate.

akshaykankal@gmail.comLinkedIn

Languages

English · German (learning) · Hindi · Marathi