Skip to content

Product Manager · Boston, MA

I understand technology, people, and business — and build products where all three create real value.

For 4 years I've turned complex B2B workflows in finance, tax, procurement, AI and field services into products that feel simple to use. I start with the question most teams skip, “should we build this?”, and carry the answer from discovery all the way to delivery.

Currently completing an MS in Information Systems at Northeastern University in Boston.

Portrait of Aditya Hasija

Portfolio index

Aditya / Product IndexSystem: Online

Index

Index: Pick a chapter.

Boston / MAPick a key
Years across B2B product & technology
4
SaaS product roadmaps owned
5
Enterprise clients supported
60+
End-to-end implementations
10+

The short version

I didn’t start out wanting to be a Product Manager. I started much closer to the users — implementing software, answering questions, hearing complaints, and figuring out why something that made perfect sense on paper didn’t always make sense to the person using it.

That experience shaped how I think about products today.

Over the years, I moved from translating customer problems into requirements to owning roadmaps, running discovery, working alongside engineers and data teams, and making decisions about what should be built, what shouldn’t, and why. Along the way, I’ve worked across finance, tax, procurement, field services, data, and AI.

What I enjoy most sits at the intersection of three things: technology, people, and business. Understanding what technology makes possible, what people actually need, and what creates meaningful business value — then finding the product opportunity where those three meet.

I’m still curious about the same thing that got me started: why people use some products, struggle with others, and love a few.

Focus
Product Manager · AI Product Manager · Product Analyst · Business Analyst
Based in
Boston, Massachusetts
more of the story in the notebook ↓

Where I've worked

  1. Massey Services, Inc. logo

    Product Analyst

    Massey Services, Inc. · Internship

    Jul 2025 – Dec 2025 · 6 mosOrlando, Florida, United States · On-site

    Worked as a Product Analyst in the field services industry, supporting digital product development and modernization across customer-facing and internal operations. Collaborated with business leaders, field teams, and technical stakeholders to translate operational challenges into practical, scalable product solutions.

  2. BDO in India logo

    Product Manager - Associate

    BDO in India · Full-time

    Oct 2022 – Aug 2024 · 1 yr 11 mosMumbai, Maharashtra, India

    Worked in product management and business analysis, developing enterprise SaaS solutions in the tax and finance domains, with a focus on regulatory compliance and process automation. Bridged business and technology by collaborating with clients, domain experts, engineers, and data scientists to shape product strategy and deliver customer-centric solutions throughout the product lifecycle.

  3. Expenzing logo

    Product Specialist - Business Analyst

    Expenzing · Full-time

    Jun 2021 – Sep 2022 · 1 yr 4 mosMumbai, Maharashtra, India

    Specialized in enterprise SaaS products for procurement, employee expense management, and strategic sourcing, serving clients across the Banking, Financial Services, and Insurance (BFSI) industry. Bridged business and technology by collaborating with clients and technical teams to translate business requirements into tailored product solutions, streamline workflows, and drive successful implementation, adoption, and ongoing client support.

Four chapters, one thread

How curiosity about technology turned into a career in product, told across four spreads.

At a glance · jump to a page

Chapter 01
younger me (photo to come)

Before product, there was curiosity.

Before I knew what a Product Manager was, I was the kid who wanted to understand how technology worked.

Computers, games, software — I was fascinated by the fact that someone could take an idea, build a system around it, and suddenly millions of people could interact with it.

Gaming probably taught me my first product lesson without me realizing it:

Great technology disappears when the experience just works.

  • How does this actually work?
  • gaming → computers → technology → ??? → product
  • CTRL + ALT + CURIOSITY
  • Hours spent gaming: classified

Pages 1 to 2 of 8: Curiosity & TSEC

Selected projects

A few things I've built to explore how products, systems, and people interact.

All projects on GitHub
Live · Release 1.2
Touchline tactics editor: a 2-3-1 shape on a pitch with a pressing trap, runs and notes
Player view on a phone: the starting lineup with one player highlighted by Find me

Featured project

Touchline

The matchday tactics board for grassroots captains and coaches, finally digital.

On a Sunday-league side the game plan lives in a WhatsApp thread and a blurry whiteboard photo. Touchline puts it in one place: pick the squad, draw the shape, and publish a briefing every player can open.

  1. Squad
  2. Fixture
  3. Lineup
  4. Tactics
  5. Briefing
  • One link for the whole team. Players open the briefing on their phone, with no app or account, and tap Find me to see just their job.
  • Three releases, three questions. A spec-first MVP; then roles, set pieces and fair minutes after I used it myself; then a UX audit of 38 screens that reworked the first-run journey.
  • Product strategy
  • PRD & PR-FAQ
  • UX audit
  • Next.js
  • Supabase
  • AI-assisted build

Featured AI project

Sillage

A fragrance collection manager with an AI advisor that can't make things up.

Collectors still buy the wrong next bottle and wear the same one every day. Sillage starts from what you own: it shows what your shelf is missing, what to wear today, and what to buy next, with prices for the exact bottle.

  • A grounded AI agent. Plans and calls ten typed tools over tested engines; anything a tool didn't return is removed before you see it. Runs free on Gemini, or on Claude.
  • Built around your collection. A Collection Score, a daily pick from bottles you own, and a lab that predicts how two of them layer.
  • Prices you can trust. Exact-bottle matching at 100% precision on adversarial listings; suspiciously cheap and marketplace offers are flagged, never hidden.
  • AI agents
  • LLM evals
  • Spec-first PRDs
  • UX audit
  • Next.js
  • Supabase
  • Gemini + Claude APIs
Live MVP · no sign-up
The Sillage shelf: six drawn bottles, a Collection Score of 82, and a card asking what to wear today
Sillage on a phone: the shelf and Collection Score with bottom navigation
Clickable Figma prototype
WorkShip job hunter home: welcome header, search, matched skills, verified featured jobs and company notifications
Sign-up screen: One Platform. Verified Jobs. Verified Talent., with a job hunting or hiring choice

Product & UX case study

WorkShip

Smarter hiring, zero noise.

Recruiters wade through fake or irrelevant applications, while strong candidates get lost and re-fill the same forms on every platform. WorkShip is a two-sided job portal where candidates and listings are both pre-verified.

  1. Verify once
  2. Quick Apply
  3. Real-time status
  • Trust on both sides. Every company carries a Trust Score, and recruiters hire from a pool of pre-verified talent.
  • Career Compass and Wingman AI. A dashboard for the whole application journey, and an assistant that drafts the follow-up email.
  • My role. On a team of five, I led product and design: problem framing, requirements and use cases, UX research (personas, empathy maps, competitive analysis against LinkedIn, Indeed and Workday), the Figma wireframes and UI, and the final presentation.
screens
~100
flows
7
on the team
5

A design case study, from sign-up to the recruiter flow, audited and rebuilt as a v2 with design tokens and components. No production code.

  • Problem framing
  • Product requirements
  • Use cases
  • UX research
  • Competitive analysis
  • Design system
  • Figma
CampusCribTeam of threev2 · rebuilt
CampusCrib's student search: filters for rent, bedrooms, distance and amenities beside illustrated apartment cards with prices
An electrician's work queue: an urgent request with its history and a Mark completed button
One network · four enterprises
  • CustomerStudents searching and booking
  • Real Estate AgencyAgents listing apartments
  • Resident AssociationResidents raising requests
  • Building ManagementManagers and service crews

Requests route toPlumbingElectricalCarpentryPest control

CampusCrib

Finding the apartment is only the start.

Student housing runs on handoffs: students, leasing agents, residents, building managers and service crews all touch the same building. CampusCrib puts the whole journey in one system, from searching and booking an apartment to the leaky tap after move-in.

  • What we built. With a team of three: a Java Swing desktop app on MySQL, modelled as enterprises, organizations and roles.
  • What v2 adds. I rebuilt it so every role works end to end: everything saved to MySQL, hashed passwords, and requests routed to the right crew with a full status history, plus reports for building management.
roles across 4 enterprises
9
tables, all persisted
6
tests on H2 and MySQL
25
  • Java 17
  • Java Swing
  • MySQL / JDBC
  • Role-based workflows
  • JUnit
  • Maven
SKILLPORTv2 · rebuilt
SkillPort course page: React Fundamentals, with the learner's progress at 40%, a Continue learning button and the lesson list with completed lessons ticked
The same course page on a phone

SkillPort

Learn from people who do the work.

A peer-to-peer learning platform: anyone can teach a short course, and learners enroll, track progress lesson by lesson and ask questions in each course's discussion. v2 rebuilds my original prototype, which never ran end to end, into a working, tested app.

  1. Browse
  2. Enroll
  3. Learn
  4. Discuss
  5. Certificate

What I built. Course authoring with drafts, lesson progress and PDF certificates, course discussions with instructor-labelled replies, and an instructor dashboard showing how learners are getting on.

roles, enforced in the API
2
screens
12
automated tests
32
  • React
  • Node.js / Express
  • MongoDB
  • JWT auth
  • Role-based access
  • Vitest

Building & coming soon

Building nowAI agents · in progress

ProductLens AI

Research smarter. Validate faster. Build better.

An AI research and decision workspace for product managers. Describe a new feature or an improvement, and it turns the research into an evidence-backed decision brief: the risks, the metrics that would show success, and the experiment to run next.

  • Agents that check each other. Market, customer and risk agents research a feature idea; a verifier flags weak evidence and separates facts from assumptions.
  • Decisions, not opinions. Every brief ends in success metrics and an experiment to run next, not a one-time verdict on whether an idea is good.
  • First milestone. One research agent producing a structured, cited finding for a feature idea, with a run trace and evals. Live analytics and real A/B tests are out of scope.
  • Multi-agent design
  • Source verification
  • Structured outputs
  • LLM evals
  • Product discovery
Concept · planned workflow

Feature idea

“One-page checkout to help more customers complete purchases”

  • Marketresearch agent
  • Customerresearch agent
  • Riskresearch agent

Verifier

Checks sources · flags unsupported claims · fact vs assumption

Decision brief

Cited findings · risks · KPIs (e.g. checkout completion, payment errors) · a test plan

Biggest risk: confident ≠ correct. Verification is designed in from day one.

Two AI products with PRDs and technical designs written, building next.

Up nextAgent infrastructure · specced

AgentOps AI

Autonomous when possible. Accountable at every step.

Agent demos hide their failures: a polished final report can't tell you whether an action was authorized, a claim was supported, or how a broken step recovered. AgentOps runs supervised research missions where every step is recorded and risky actions wait for a human.

Concept · mission runevent log ●
  1. Planner
  2. Researcher
  3. Analyst
  4. Reviewer⏸ approval
  5. Report

tool.requested export_report

policy.evaluated effect: approval

approval.requested waiting for operator…

  • Permissions live in the backend. Prompt text never grants access; a denied action is never executed.
  • Decision Replay. Rebuild any run from its recorded events: which evidence, tool result and policy came before each action.
  • Scoped on purpose. One research workflow and a read-only MCP tool first; no payments, email or open web actions.
  • LangGraph
  • MCP
  • Human-in-the-loop
  • Event sourcing
  • FastAPI
  • React Flow
Up nextLLM evaluation · specced

EvalForge AI

Find how AI fails before users do.

An average quality score can go up while a critical failure slips through. EvalForge is a failure-first workbench for document-grounded assistants: run the old and new versions on the same pinned cases, fingerprint what breaks, and get a go/no-go a PM can defend.

Concept · release checksame pinned cases

Baseline

Prompt v1

Candidate

Prompt v2

  • Unsupported claim
  • Missing citation
  • Injection compliance
  • Contradicted evidence
GatePassConditionalFail1 critical case
  • Critical failures override averages. One severe regression fails the release, even if the mean score rises.
  • Failures become test cases. Each one is saved with its exact prompt, evidence and scorer version, so it reproduces without a new model call.
  • Judges are fallible. Deterministic checks first, rubric-based LLM judging second, human labels to calibrate.
  • LLM evals
  • LLM-as-judge
  • Adversarial testing
  • Release gates
  • FastAPI
  • Next.js

Capabilities & Toolkit

Product thinking, technical fluency, data, and delivery — the tools I use to move from ambiguity to execution.

Product & Strategy

Deciding what to build and why: roadmaps for 5 SaaS products, shaped by 40+ customer interviews.

  • Product Strategy
  • Product Discovery
  • Roadmapping
  • Feature Prioritization
  • Backlog Management
  • PRDs
  • User Stories
  • Acceptance Criteria
  • Go-to-Market Thinking
  • Competitive Analysis
  • Market Research
  • Product Lifecycle Management
  • KPI Definition
  • Experimentation
  • Customer Discovery
  • Voice of Customer
  • Product Requirements
  • MVP Scoping
  • MoSCoW Prioritization
  • Stakeholder Alignment

AI & Technical

Enough technical fluency to prototype, spec APIs and work closely with engineering and ML teams.

  • AI Product Thinking
  • Prompt Engineering
  • LLM Workflows
  • Generative AI
  • AI Prototyping
  • API Integrations
  • REST APIs
  • JSON
  • System Thinking
  • Technical Requirements
  • Basic System Design
  • Data Flows
  • Git
  • GitHub
  • React
  • Next.js
  • Java
  • JavaScript
  • HTML/CSS
  • Python
  • R
  • SQL Databases
  • Claude Code
  • v0
  • Vercel
  • Netlify

Data & Analytics

Defining the metrics that matter, and building the dashboards that track them.

  • SQL
  • Data Analysis
  • KPI Tracking
  • Dashboarding
  • Data Visualization
  • Excel
  • Power BI
  • Tableau
  • Google Sheets
  • Descriptive Analytics
  • Trend Analysis
  • Data-driven Decision Making
  • Basic Statistical Analysis
  • Reporting
  • Business Intelligence
  • Python
  • R
  • Snowflake
  • Databricks
  • Alteryx

Business Analysis

Turning messy, real-world workflows into requirements engineers can build against.

  • Requirements Gathering
  • BRD
  • SRS
  • Functional Requirements
  • Non-functional Requirements
  • Process Mapping
  • Gap Analysis
  • Impact Analysis
  • Workflow Design
  • Use Cases
  • UAT
  • Root Cause Analysis
  • Business Process Improvement
  • Stakeholder Interviews
  • Requirements Traceability
  • API Requirements
  • Solution Analysis
  • Documentation
  • Process Optimization

Delivery & Program Management

Getting it shipped: sprints, releases, client implementations and pilots across teams.

  • Agile
  • Scrum
  • Sprint Planning
  • Backlog Grooming
  • Cross-functional Delivery
  • Project Planning
  • Program Coordination
  • Risk Management
  • Dependency Management
  • RACI
  • Release Planning
  • Implementation
  • Change Management
  • Training
  • Client Onboarding
  • Stakeholder Management
  • Status Reporting
  • Roadmap Execution
  • Pilot Planning

Tools & Platforms

The day-to-day stack for planning, design, documentation and building.

  • Jira
  • Confluence
  • Figma
  • Miro
  • Lucidchart
  • Microsoft Visio
  • Excel
  • PowerPoint
  • Power BI
  • Tableau
  • GitHub
  • VS Code
  • NetBeans
  • Postman
  • Vercel
  • Netlify
  • Claude Code
  • v0
Additional learning
  • Business Analysis Fundamentals (ECBA, CCBA, CBAP)Udemy · 2023
  • Software ArchitectureUdemy · 2023
  • Data Analyst BootcampUdemy · 2023

What People Say

The people I've worked with usually describe me better than I describe myself.

…He consistently demonstrated exceptional technical expertise, taking ownership and initiative in enhancing our product suite. Aditya was always bursting with innovative ideas, constantly brainstorming ways to improve our solutions.

Nishant Dalvi

Managed Aditya directly, Product Strategy · BDO India

…Aditya excels at translating complex requirements into actionable solutions and effectively communicating these to cross-functional teams.

Manan Shah

Dev Lead, worked with Aditya · BDO India

Blog — Coming Soon

Notes on products, technology, and the messy parts in between.

I'm putting together a space to write about product decisions, AI, business problems, experiments, and things I learn while building.

  • 01Product
  • 02AI & Technology
  • 03Building & Experiments

Let's build something useful.

I'm always happy to talk about products, technology, interesting problems, or opportunities to build something that matters.