Portfolio Manager
Tracking What Bridgewater and Renaissance Are Buying β 13F Filings Decoded
Key Takeaway
My AI agent monitors SEC 13F filings from the top 20 hedge funds, extracts new positions, exits, and sector shifts, and delivers the intel within 24 hours of filing β before financial media picks it up.
The Problem
Institutional investors managing billions of dollars are required to disclose their holdings quarterly via SEC 13F filings. It's one of the few legal edges available to retail investors: seeing what the smartest money in the world is actually buying.
The problem is timing and processing.
13F filings drop on EDGAR as XML files. They're dense, poorly formatted, and there are thousands of them. By the time Bloomberg or CNBC writes an article about what Bridgewater bought, the market has already priced in the information. You're reading about it days or weeks after the filing.
And even if you catch the filing early, comparing it against the previous quarter's filing β figuring out what's new, what increased, what got dumped β requires a spreadsheet and patience.
I wanted the opposite: an automated system that watches the filings, parses them instantly, compares quarter-over-quarter, and tells me what matters. Before the headlines.
The Solution
The SEC Filing Watcher agent on Mr.Chief monitors EDGAR's XBRL feed for 13F filings from a curated list of 20 hedge funds. When a new filing drops, the agent parses the holdings, diffs against the previous quarter, and delivers a structured alert via Telegram.
It also cross-references against my own holdings β flagging when a fund I track buys or sells something I own.
The Process
yamlShow code
# SEC 13F Watcher
name: 13f-filing-watcher
schedule: "0 */6 * * *" # Check every 6 hours during filing season
channel: telegram
task: |
Monitor SEC EDGAR for new 13F filings from tracked funds.
TRACKED FUNDS:
- Bridgewater Associates
- Renaissance Technologies
- Citadel Advisors
- Two Sigma Investments
- DE Shaw & Co
- Millennium Management
- Point72 Asset Management
- Tiger Global Management
- Pershing Square Capital
- Appaloosa Management
- Baupost Group
- Third Point LLC
- Elliott Investment Management
- Viking Global Investors
- Coatue Management
- Dragoneer Investment Group
- Lone Pine Capital
- Maverick Capital
- Greenlight Capital
- Icahn Enterprises
On new filing:
1. Parse all positions
2. Diff vs previous quarter
3. Categorize: NEW | INCREASED | DECREASED | EXITED
4. Rank by position size change ($)
5. Cross-reference with MY holdings
6. Identify sector rotation themes
7. Deliver structured alert
The parsing engine:
pythonShow code
async def process_13f_filing(fund_name: str, filing_url: str):
# Parse the XBRL/XML filing
current = parse_13f_xml(await fetch(filing_url))
previous = load_previous_quarter(fund_name)
# Diff
changes = {
"new_positions": [],
"increased": [],
"decreased": [],
"exited": [],
}
current_tickers = {h["cusip"]: h for h in current["holdings"]}
previous_tickers = {h["cusip"]: h for h in previous["holdings"]}
for cusip, holding in current_tickers.items():
if cusip not in previous_tickers:
changes["new_positions"].append(holding)
elif holding["shares"] > previous_tickers[cusip]["shares"]:
holding["change_pct"] = (
(holding["shares"] - previous_tickers[cusip]["shares"])
/ previous_tickers[cusip]["shares"] * 100
)
changes["increased"].append(holding)
elif holding["shares"] < previous_tickers[cusip]["shares"]:
changes["decreased"].append(holding)
for cusip in previous_tickers:
if cusip not in current_tickers:
changes["exited"].append(previous_tickers[cusip])
# Cross-reference with my portfolio
my_holdings = load_my_portfolio()
overlaps = find_overlaps(changes, my_holdings)
# Sector analysis
sectors = analyze_sector_rotation(changes)
return format_13f_alert(fund_name, changes, overlaps, sectors)
A real alert:
View details
π¦ 13F FILING ALERT β Renaissance Technologies
Filed: Mar 11, 2026 (Q4 2025 holdings)
π₯ NEW POSITIONS (top 5 by value):
1. PLTR β 2,100,000 shares ($178M) β YOU HOLD THIS
2. COIN β 890,000 shares ($201M)
3. RKLB β 3,400,000 shares ($95M)
4. HOOD β 2,800,000 shares ($87M)
5. AFRM β 1,200,000 shares ($72M)
π INCREASED (top 5):
1. NVDA β +1,800,000 shares (+12%) β YOU HOLD THIS
2. GOOGL β +2,400,000 shares (+8%) β YOU HOLD THIS
3. MSFT β +900,000 shares (+3%)
4. AMZN β +1,100,000 shares (+5%)
5. META β +700,000 shares (+4%)
π DECREASED (top 3):
1. AAPL β -3,200,000 shares (-15%)
2. JPM β -1,800,000 shares (-20%)
3. XOM β -2,100,000 shares (-25%)
πͺ EXITED:
1. INTC β full exit (was 4.2M shares)
2. T β full exit (was 1.8M shares)
π― PORTFOLIO OVERLAP:
PLTR: Renaissance initiated β confirms your thesis β
NVDA: Increased position β aligned with you β
GOOGL: Increased position β aligned with you β
TSLA: No change β neutral signal
π SECTOR ROTATION:
β² Technology (especially AI/data infra)
β² Fintech / Crypto-adjacent
βΌ Energy, Telecoms, Traditional Banking
The Results
| Metric | Before | After |
|---|---|---|
| Time to analyze a 13F filing | 1-2 hours | Instant |
| Time from filing to my awareness | 2-7 days (media delay) | <24 hours |
| Funds tracked | 0 (occasional article) | 20 |
| Quarter-over-quarter diff | Manual spreadsheet | Automated |
| Portfolio overlap analysis | Never did this | Every filing |
| Sector rotation signals per quarter | 0 | 4-6 |
| Conviction boost on holdings | Anecdotal | Data-backed |
The PLTR conviction signal alone was worth the setup. Renaissance initiating a position at $85 while I was holding at $72 average β that's not just confirmation bias. That's one of the most sophisticated quant funds in history putting money behind the same thesis.
Try It Yourself
- Set up the SEC Filing Watcher skill in Mr.Chief (EDGAR API is free, no key needed)
- Curate your fund list β focus on funds whose strategy aligns with your investing style
- Input your portfolio for automatic overlap detection
- During filing season (45 days after quarter end), set to check every 6 hours
- Between seasons, reduce to weekly checks for amendments and late filings
The information edge from 13F filings isn't the data itself β it's public. The edge is processing speed and structured comparison.
When Renaissance buys what you already hold, it's not luck. It's conviction, confirmed.
Related case studies
Portfolio Manager
Insider Buying at PLTR β The Agent Alerted Me 6 Hours Before CNBC
AI agent monitors SEC Form 4 filings in real-time, alerting on insider purchases hours before financial media coverage. Information edge through autonomous filing surveillance.
Portfolio Manager
The Complete Financial Operating System β All 17 Skills in One Stack
How 17 AI finance skills work as one integrated system: data ingestion, analysis, execution, and reporting. 31 agents across 8 teams running a complete financial operating system on Mr.Chief.
Portfolio Manager
Earnings Preview Report for 5 Holdings β Before the Market Opens
How an AI agent generates earnings preview reports for NVDA, GOOGL, META, TSLA, and PLTR in 10 minutes β consensus estimates, implied moves, and key metrics delivered before Monday open.
Want results like these?
Start free with your own AI team. No credit card required.