Humans are Search Engines and Agentic Search connects fuzzy knowledge graphs

Human are query-able search engines. Text in, Text out. 

Each search engine is a fuzzy knowledge-graph.

Agentic Search can connect to many different search engines in an agentic loop for deep research. Including humans.

I need 400ckal/h on my bike

 

Calculated by Gemini, not double checked. Seems low.

Gravel Cycling Energy Model

Speed (km/h):

Rider Weight (kg):

---

Caloric Consumption: 0.0 kcal/h

Total Power: 0.0 W

Breakdown:

• Aerodynamic Power: 0.0 W
• Rolling Resistance: 0.0 W

Just use OpenSearch

1. LLM-Query Compatibility & Representational Density

For autonomous agents, "ergonomics" are a human distraction. The primary metric is Representational Density in LLM training corpora. OpenSearch (ES 7.10 fork) utilizes a JSON-based DSL that is the most documented search interface in history.

• Zero-Shot Reliability: Agents generate nested bool queries (must/filter/should) with significantly lower hallucination rates compared to Vespa’s YQL or Solr’s XML-adjacent syntax.
• Deterministic Error Handling: Structured JSON error responses allow agents to parse stack traces and auto-correct query syntax in multi-stage reasoning loops.

2. Lexical Primacy & Late Interaction (ColBERT)

Medical IR demands exact-match precision for biochemical entities. OpenSearch provides unadulterated BM25 control, avoiding the "black-box" typo-tolerance found in vector-first databases.

$$score(D, Q) = \sum_{q \in Q} IDF(q) \cdot \frac{f(q, D) \cdot (k_1 + 1)}{f(q, D) + k_1 \cdot (1 - b + b \cdot \frac{|D|}{avgdl})}$$

The ColBERT Advantage: Unlike standard bi-encoders that compress abstracts into a single vector, OpenSearch 3.x supports multi-vector Late Interaction. Using the MaxSim operator, the engine preserves token-level nuances (e.g., "Inhibitor X" vs. "Protein Y") that are often lost in 1536-dimensional averages.

3. Single-Node Batch Efficiency

With a 12M document corpus and monthly batch updates, we optimize for Read-Heavy Static Segments over real-time mutability.

Metric	OpenSearch (Lucene)	Vespa (C++)	Manticore (SQL)
Memory Strategy	OS Page Cache + 32GB Heap	Tensors / Mmap	Columnar / Disk
Latency (Agentic)	< 5s (Complex Hybrid)	< 1s (High Throughput)	< 2s (SQL Joins)
Lindy Effect	High (Established Standard)	Medium (Enterprise-Niche)	High (Sphinx Heritage)

By setting index.refresh_interval: -1 and index.number_of_replicas: 0 during ingestion, OpenSearch builds contiguous Lucene segments that maximize hardware utilization without the overhead of distributed consensus.

4. Licensing Stability & Governance

In a landscape of "corporate rug-pulls," OpenSearch (Linux Foundation) provides the highest resistance to licensing shifts. Unlike venture-backed alternatives (Weaviate/Typesense) or commercial-pivots (Vespa.ai), OpenSearch remains a community-governed Apache 2.0 utility.

OpenSearch is the optimal choice because it treats code as a liability and cognitive efficiency as a priority. It offers the best blend of Lexical Rigidity, Agent Compatibility, and Operational Insurance.

Putting myself in a suit

A useful usecase for nano banana. It is nice to iterate on my style like iterting on source code. This could be great for the next haircut, suit purchase, etc.

With clear marking and a nice fitting quote in the age of ai.

Other iterations:

Allgäu real estate vs s&p the last 30 years

 

I hear again and again, that the houses here gained so much value over the last 30 years. That may be true, but nobody thinks about the opportunity costs. Also, exponentials are non-intuitive.

We start with 100k 30 years ago. s&p returns 2mio, 22x return, real estate 876k, so 8.8x return (assuming you re-invest rental income. If not 375k or 3.7x).

Lets hope the s&p keeps going.

Year,SP500_EUR_Wealth,House_Wealth_TCO,House_Price_Only
1995,100000.0,90090.09009009009,90090.09009009009
1996,120259.99999999999,90348.64864864864,89189.18918918919
1997,157552.626,90135.48648648648,87851.35135135135
1998,199571.91135419998,90821.16628378378,87412.09459459459
1999,238548.30564167525,91957.52351351352,87412.09459459459
2000,214359.50744960937,93559.35014695946,87849.15506756755
2001,186407.0276781803,94227.88221702365,87409.90929222971
2002,142843.70530978957,94410.56892744442,86535.81019930741
2003,180525.87477051208,93624.8237708347,84805.09399532127
2004,196755.15091238113,93306.38064288226,83533.01758539144
2005,202657.80543975256,93448.38677277742,82697.68740953752
2006,230259.79854064688,94523.4567091014,82697.68740953752
2007,238387.96942913168,96554.51191187964,83524.6642836329
2008,146632.4399958589,99593.13919851821,85195.15756930556
2009,181017.7471748878,101204.17962815377,85621.13335715208
2010,204151.81526383845,105386.77199265065,88189.76735786664
2011,204151.81526383845,111859.90091671808,92599.25572575998
2012,231528.5736907192,120412.83117182789,98618.20734793438
2013,300061.0315031721,130821.98295740236,106014.57289902946
2014,334237.98299138347,142776.18619749692,114495.73873095182
2015,331798.04571554635,156527.12441908423,124227.87652308273
2016,363451.57927680947,172374.8746271339,135408.38541016018
2017,434033.87597236584,190678.0260830252,148272.1820241254
2018,406950.1621116902,211866.12089427272,163099.40022653795
2019,524477.3689295463,233245.19027596732,177778.34624692638
2020,609757.3891174906,255578.59502323743,192889.50567791512
2021,773721.1510511838,286475.63604272594,214107.3513024858
2022,623309.7592868337,300829.3928740446,222671.64535458523
2023,774337.7139620335,279426.1943225619,204857.91372621842
2024,954835.8350865834,276447.5602569827,200760.75545169404
2025,1111333.4284572743,295800.897082526,212806.4007787957

How whey is made

Source: https://www.youtube.com/watch?v=LeUZmojH7p8

1. Start with Milk, homogenize, pasteurize. It is 87% water, 5% lactose, 3.5% fat, 3.5% protein of which 80% is casine and 20% whey. So there is 0.7% whey in milk. Use around 10l milk for 1kg cheese.
2. Whey is a byproduct from making cheese which is made from the casine protein. You need to sperate that caseine from the rest of the milk. Add acid or bacteria that eat lactose and secret acid. Curdling happens and the milk gets firm as the casine separates from the rest, including the whey protein which stays in the water.
3. Now it is one big curd with all the other milk components trapped inside. Cut it in small pieces an press it. What comes out are the curds for the cheese and whey liquid.
4. The whey liquid contains protein, lactose and fat. 
5. Filter out the fat, lactose and minerals. Then dry it. You get whey

In short: Milk -> Filter out casine via curds -> filter out fat, minerals and lactose -> dry it -> whey powder.

smartphone holder bicyle