v1.0.0 // system online
Suraj Phanindra
>
"AI capability isn't the bottleneck. Context is."
"AI capability isn't the bottleneck. Context is."
I'm Suraj, an AI/ML engineer and co-founder based in Sunnyvale, California. I build enterprise AI systems that actually work in production, serving Fortune 500 companies with RAG platforms processing 100,000+ documents per deployment.
My journey spans from winning a Math Olympiad at age 7 to architecting knowledge systems for multi-billion-dollar enterprises. What hasn't changed: the belief that building something meaningful matters more than just getting back up after falling.
Currently, I'm the Co-Founder & Chief Architect at CompanyInsights.AI, where I design agent architectures and knowledge systems. When I'm not building, I'm researching CoT faithfulness, exploring quantitative finance, or mixing music.
500K+
End Users Served
95%+
RAG Accuracy
Fortune 500
Enterprise Clients
“Not everything needs an LLM.”
Precision engineering means choosing the right tool, not the trendiest one.
“For now, not forever.”
Ship to learn, keep options open. V1 is never the last version.
“Context is the bottleneck.”
AI capability isn't the bottleneck. Context is. The solution is always knowledge architecture.
“Build, don't just recover.”
Falling down is okay. Getting back up is important. Building something after is what matters.
// projects
From startup MVPs to Fortune 500 enterprise deployments
Enterprise AI That Knows Your Business
Enterprise RAG platform that transforms how Fortune 500 companies interact with their institutional knowledge.
100,000+
Docs Per Deployment
95%+
Retrieval Accuracy
1M+
Vector Embeddings
51 files
Agent Architecture
Leetcode is so 2019
Agentic technical interview platform built for the era of AI-assisted development. Tests what really matters, not theatrics and memorization.
3
AI Agents
Real-time
Evaluation
Cloud Sandbox
Platform
One Memory. Every AI.
Open source browser extension that unifies context and memory across every AI platform you use.
3,000+
Pre-release Impressions
Cross-platform
Platforms Supported
Open Source
License
AI Brand Intelligence Platform
Monitor how LLMs perceive your brand across AI-generated answers. Runs queries against multiple providers, extracts mentions, detects comparisons, produces probabilistic visibility metrics.
4
LLM Providers
30
Runs Per Query
Wilson Score
CI Method
Natural Language Vehicle Search
Natural Language Vehicle Inventory Search Platform for used car dealerships.
Natural Language
Search
Web App
Platform
Attendance, Reimagined
All-in-one access credential application providing seamless check-in, foot traffic data, and customer identity validation using BLE beacons and mobile app.
Wadhwani IGNITE Winner
Award
Provisional Patent
IP
5 Remote Engineers
Team
Wadhwani Foundation
Funding
When ML Meets Markets
Quantitative trading strategy using machine learning to achieve superior risk-adjusted returns, developed as an IIQF thesis.
>1.0
Sharpe Ratio
92%
Drawdown Reduction
131
Features Analyzed
// journey
SFO International Math Olympiad: All India Rank 2, International Rank 43 (Grade 2)
National Science Talent Search Exam Gold Medalist. Scored 95% in ICSE Board exams.
Scored 2400/2400 on the SAT. Top 1% performer in India.
Admitted to BITS Pilani (1.47% acceptance rate) for B.E. Computer Science.
Co-founded AirStamp. Won Wadhwani IGNITE. Filed provisional patent.
Licensee & Head Organizer at TEDxBITSPilani.
Joined Webless.ai. Built Search and Retrieval systems, Founding team 0-1, $9M Valuation
Presented at Web Summit Lisbon 2024. Systems serving 500K+ users.
Completed AI Trading Strategy thesis. Sharpe ratio >1.0, 92% drawdown reduction.
Joined as Founding Engineer at CompanyInsights.AI.
Launched Singularity: cross-platform AI memory. 3,000+ pre-release impressions.
Promoted to Co-Founder & Chief Architect at CompanyInsights.AI.
Built 51-file agent knowledge architecture (~14,750 lines). Fortune 500 deployments.
// thoughts
Where I write about AI, engineering, and building in public
Coming Soon
Everyone's racing to build bigger models. I'm more interested in what happens when the model is smart enough but doesn't know what you need.
Coming Soon
Sometimes a regex, a lookup table, or a well-designed rule engine will outperform any LLM. Here's how I decide what tool fits.
Coming Soon
51 files. 14,750 lines of knowledge infrastructure. Here's the framework I built to prevent agents from writing plausible but wrong code.
// connect
Find Me Online
GitHub
suraj-phanindra
--
Repos
--
Stars
--
Contributions
Check out my open-source contributions and side projects
LinkedIn
Suraj Phanindra
Co-Founder & Chief Architect at CompanyInsights.AI
Thought leadership, technical insights, and building in public
YouTube
@SurajPsamdude
Technical deep-dives and AI engineering content