available now · 16+ yrs · Hyderabad, India (IST)

Abhishek Maurya

Staff Software Engineer | AI agent platforms, MCP and evals

Results

95%
less research and debugging effort for 20+ teams on MCP workflows (measured across the 7 pilot squads in the 4-week rollout)
−90%
case task resolution time on the Agentic Developer Platform
20% → 90%
legacy test coverage lift on ADP pilot squads
single digits
open external defects org-wide, down ~90%
80% → 95%
accessibility compliance lift, Angular-to-React migration at Reputation
9
engineers led and mentored as tech lead at ServiceNow
All 9 results, with the work behind them: work page →

Selected work

3 shipped agent systems. Each one has a deep dive.

run 01 · ServiceNow✓ shipped

Agentic Developer Platform (ADP)

Creator, ServiceNow

Multi-team bootstrap and governance infrastructure for Claude Code, MCP servers, agents, and skills. Built in four weeks on a multi-agent orchestration pattern, rolled out to 7 pilot squads across 20+ teams with 20+ skills and 20+ agents.

−90% task resolution time · 20% → 90% legacy coverage · open defects → single digits org-wide · Days → hours · Multi-agent pattern
run 02 · ServiceNow✓ shipped

MCP-driven engineering workflows

20+ teams, ServiceNow

Integrated MCP-driven agentic workflows into the engineering lifecycle organization-wide. Reduced research and debugging effort by approximately 95% while improving accuracy.

−95% research & debugging · 20+ teams · Higher accuracy
run 03 · ServiceNow✓ shipped

Build Agent, ServiceNow Studio

App Engine, ServiceNow

Elevated App Engine with agentic workflows for AI-assisted application creation, flow generation, and tool orchestration across 35+ metadata types and 11 domains. Built on the planning pattern (think first, then execute) with a metadata-aware, multi-model (Claude default; Now LLM + Claude on Bedrock), self-healing agentic architecture.

35+ metadata types · 11 domains · Multi-model · Self-healing · Metadata-aware · Planning pattern (think → execute)
All 4 case studies, with metrics and run logs: work page →

Selected builds

Side projects, shipped and living online.

SpendIQ Copilot

Agent

14-case golden-set eval harness passes with majority voting: agent answers enterprise data questions without inventing numbers

Conversational spend-analysis agent: LangGraph ReAct over SQLite + FTS5 with SQL guardrails (SELECT-only, read-only connections), an AST-whitelisted calculator instead of eval, and a 14-case golden-set eval harness with majority voting, all passing

Visit SpendIQ Copilot →

ordo

iOS app

Time-to-first-token cut from 30s to 805ms on an iPhone A16 by changing architecture and resolution, not hardware

On-device vision-language model built on Qwen2-VL-2B with llama.cpp across a GGUF quantization ladder, deployed to iPhone (A16), with a private photo eval set, per-stage TTFT decomposition, and LoRA accuracy recovery

Visit ordo →
Screenshot: Revision

Revision

Desktop app

Principal-level interview prep that works fully offline from a single SQLite file: no account, no sync, no network

Local-first Tauri desktop app for principal-level interview prep: FSRS-5 spaced repetition, drag & air gestures, single SQLite file, fully offline. DSA / system design / AI / behavioral decks with glassmorphic review, cloze + KaTeX, and 53-week analytics

Visit Revision →
Screenshot: Switchyard

Switchyard

Simulator

Every internal step of an agentic RAG loop (ingestion, retrieval, tool calls, memory) is watchable instead of imagined

Interactive simulator for an agentic AI + RAG pipeline: watch ingestion, retrieval, tool calls, memory, and a live trace of every internal step while the loop runs.

Visit Switchyard →
5 more builds, plus discontinued projects: work page →

What I do

8 areas, from agent platforms to backend ownership. Expand a tile for the short version.

Agentic AI & MCP MCP · Claude Code · Build Agent · Tool orchestration · Self-healing loops

Designed and shipped MCP-driven agentic workflows integrated into the engineering lifecycle, adopted by 20+ teams with roughly 95% less research and debugging effort.

Local LLM inference Ollama · MLX · turbo-fieldfare · Apple silicon · Memory-efficient inference · llama.cpp · GGUF quantization · On-device inference

Run MLX-optimized models (Qwen, Gemma, DeepSeek) locally on Apple silicon via Ollama, plus memory-efficient loaders like turbo-fieldfare that stream model portions on demand: Gemma 4 26B under 2GB RAM. Took VLMs end-to-end on-device: Qwen2-VL-2B on llama.cpp/GGUF via Metal across a quantization ladder on iPhone, with sub-second TTFT from architecture and resolution levers.

Developer platforms Multi-team · Governance · Telemetry

Founded the Agentic Developer Platform: multi-team bootstrap and governance infrastructure for agents, skills, and MCP servers.

Frontend architecture React · TypeScript · Low-code

Architected scalable React and TypeScript systems powering low-code tools like Table Builder and Workspace Builder.

Backend & data ownership Python · FastAPI · SQLite · Java / Glide · REST APIs

Hands-on backend behind the surfaces: Python/FastAPI and SQLite services with guarded, read-only SQL access, plus Java/Glide enterprise case backends and REST APIs.

System design Scalable systems · Design patterns

Metadata-aware, multi-model architectures and design patterns built to hold up across teams, tenants, and releases.

Quality & reliability Jest · RTL · Playwright · WCAG/ARIA · Performance

Enterprise testing strategy with Jest, React Testing Library, and Playwright; performance optimization and WCAG/ARIA accessibility. Lifted legacy coverage from 20% to 90%.

Technical leadership Hiring · Mentoring · Reviews

Lead a 9-person engineering team, run technical hiring and high-bar architectural reviews, and mentor engineers in AI-first development practices.

Full write-ups: work page →

Experience

16+ years across AI platforms, frontend, backend and founding teams.

ServiceNowOct 2023 – Aug 2026
Staff Software Engineer · Hyderabad
RipplingOct 2022 – Aug 2023
Senior Software Engineer · Bengaluru
ReputationSep 2020 – Sep 2022
Lead Engineer · Hyderabad
Full timeline and skills: work page → · CV →