Hacker News

Subscribe to Hacker News feed
Hacker News RSS
Updated: 37 min 55 sec ago

World Columbian Exposition in Chicago

Sun, 06/15/2025 - 7:32pm
Categories: Hacker News

A chat with Gemini AI that turns whacky

Sun, 06/15/2025 - 7:29pm
Categories: Hacker News

Ask HN: Any enterprises experimenting with AI agents / MCP-style infra?

Sun, 06/15/2025 - 7:26pm

Hi HN,

I've been building Ninja.ai solo for the past few months - a platform for deploying and observing “MCP servers,” which are essentially open protocol endpoints that AI assistants like ChatGPT or Claude can call to trigger real actions: APIs, workflows, database updates, etc.

It's all early, but I've shipped:

• A basic MCP hosting platform (Rails + Deno Deploy for isolated execution) gateway and deployment model using Deno Deploy • A gateway that aggregates multiple mcv tools under a single server allowing you to add Ninja once to your ai and then one-click install further tools • A CLI that lets developers package APIs and tools into callable MCP tools • A live app store with installable tools + observability/logging (GUI not exposed for logging yet) • An interface for agents to chat with and use these remote MCPs, even without OpenAI/Claude (max-tier) paid accounts I'm now trying to understand the real-world pain points inside enterprises that are experimenting with agents, tool-use, or "AI infra" more broadly.

If you're working at a company doing this - or know someone who is - I'd love to talk. Not to pitch, just to learn:

• What's breaking? • What's hacky? • What's needed to make this stuff production-grade? If you've shipped something similar internally (or even ruled it out), I'd really appreciate your perspective. Comment here or email's in my profile.

Happy to share what I've built so far or help troubleshoot agent infra if that's helpful too.

Thanks,

Marcus

Comments URL: https://news.ycombinator.com/item?id=44285616

Points: 2

# Comments: 0

Categories: Hacker News

EVO2

Sun, 06/15/2025 - 7:01pm
Categories: Hacker News

Show HN: Personalized Wealth Management – Institutional Meets Consumer

Sun, 06/15/2025 - 6:58pm

Problem:

If you have less than $100k to invest, you get a robo-advisor that asks you 5 questions and dumps you into one of three cookie-cutter portfolios.

If you have more than $100k, you get a human advisor who charges 1-1.5% annually to... basically do the same thing with a smile and calming voice attached.

Meanwhile, institutional investors get custom strategies built around specific durations, target dates, tax situations and actual investment goals. Not because the math is harder—but because the economics only work at scale. Here's the thing: Both traditional advisors and robo-advisors maximize profit by minimizing choice and directing capital into the bias strategies that generate them additional margins. Both just tweak a risk slider and call it "personalization." But institutional-grade portfolio construction doesn't have to be exclusive to the wealthy. The road was paved by platforms like Plaid, brining API connectivity—platforms and asset aggregation into the mainstream. Modern AI completes the picture by making true personalization economically viable via "micro-advise".

No asset transfers, no new custodians, just sophisticated strategies based on your financial goals executed where you already invest coupled with personalized financial planning & budgeting.

Technical Solution:

We've built our MVP wealth management platform that creates truly personalized portfolios by combining institutional capital market expectations stemming 30+ global asset classes. All available through low-fee publicly available ETFs. Our approach:

- SEC licensed & compliant Registered Investment Advisor - Generates unlimited unique portfolio combinations optimized for risk, return & goal specifics.

- Personalizes to individual goals, not generic risk buckets.

- Learns and improves from every user interaction - Provides institutional-grade sophistication without human bottlenecks

- Removes manager bias for in-house strategies - Uses a "glidepath" approach similar to the US retirement target-date structure to maximize achievement certainty of important life goals (down-payment, retirement, etc)

- Seeks to bring elements of habit forming platforms (like Duolingo) into retail wealth. Business Model Innovation:

-Non-custodial + AI architecture enables subscription pricing ($10/month) instead of AUM fees. Users keep control of assets while getting personalized institutional strategies.

Research Validation:

- Glidepath strategies delivered higher values in 76% of scenarios (T. Rowe Price)

- Global diversification outperformed domestic-only in 96% of 3-year periods (Hübner) - Chance of success metrics for significant life goals like retirement & major milestones are measurably improved via behavioral advantages & sequence risk protection (T. Rowe Price).

Early Results:

-Alpha users report 90%+ cost reduction vs. traditional platforms with superior personalization. Institutional style portfolios achieving goal-specific optimization that would cost minimum 10x elsewhere.

-Base model portfolios have outperformed comparable portfolios from existing market incumbent robo-platforms on both an absolute & risk adjusted basis in H1 2025.

What's Different:

This isn't another robo-advisor using basic mean reversion. It's personalization that helps you understands and discover your specific goals and adapts continuously. Think "personal wealth manager in your pocket" rather than "generic portfolio assignment." All that, in a consumer product platform designed to empower retail investors and keep them engaged.

Next Steps:

Currently in invite-only alpha at www.fulfilledwealth.co. We focused early on the portfolio construction & delivery process and are now building out the consumer-facing aspects of the web application.

Looking for feedback from the HN community on our approaches to financial personalization.

Comments URL: https://news.ycombinator.com/item?id=44285508

Points: 3

# Comments: 2

Categories: Hacker News

Pages