Marty Jacobs

Backup coding agent (to Claude Code) in 6 simple steps. Or if you want to work with coding agents on par with the costly coding agents, it's for you. Claude Code token limitations can become a huge blocker. The 5-hour token limitation surely pisses me off. So, let's fix it. Shazzadul Haque, we were talking about this. I am using opencode. Took the time today to install it; the steps aren't too complicated. I will run further experimentation to come up with workflows where I can pick up work with opencode that Claude Code was doing. Maybe, if I'm able to mirror Claude's code features (skills, agents, etc.), I might start using opencode as my main agent, who knows...we'll see. Steps to get your opencode running (I used Mac): 1. Install ollama: https://ollama.com/ Or if you have Homebrew installed: brew install ollama 2. Install open code: curl -fsSL https://lnkd.in/g72wTMZy | bash 3. Download and run the LLM model for Ollama ollama run qwen3-coder Keep it running in terminal 4. configure opencode to use Ollama model - create file: ~/.config/opencode/opencode.json - paste: {  "model": "ollama/qwen3-coder",  "provider": {   "ollama": {    "npm": "@ai-sdk/openai-compatible",    "options": { "baseURL": "http://localhost:11434/v1" },    "models": { "qwen3-coder": { "name": "Qwen3 Coder 30B" } } } } } 5. Verify installation verify ollama service: curl http://localhost:11434/api/tags verify model installation: curl http://localhost:11434/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "qwen3-coder", "messages": [{"role": "user", "content": "Say \"OpenCode ready\""}], "max_tokens": 20 }' verify opencode installation: opencode --version 6. run opencode On your terminal, run: opencode Congrats! Now you have a free coding agent that's AMAZING in your ends. Make magic happen! Good luck! Let me know how you find opencode. sidenote: If you're using Windows for all of that, you might want to look into WSL(Windows Subsystem for Linux). This will make your life much easier.

2

1 Comment

AI x asset management = $6M Multiplier, an AI company developing agent harnesses for asset managers, announced it has raised $6 million in a seed financing round. The round was led by Lux Capital, with participation from Y Combinator, GoAhead Ventures, Rebel Fund, General Advance, Unpopular Ventures, and Amino Capital. Notable individual backers include Bridgewater Associates co-CIOs Greg Jensen and Karen Karniol-Tambour, Fortress Investment Group Chairman Pete Briger, DeepMind CSO Jasjeet Sekhon, and Polygon CEO Sandeep Nailwal, along with customers such as Mercator Partners. Formerly known as WithAI, the company rebranded to align with its core mission of expanding the capabilities of human investors. The company was founded by CEO Ian McInnis, former Bridgewater investor; COO Ryan Winkler, formerly of StepStone Group; and CTO/President Ben Finch, previously a founding researcher at Sentient Labs. Multiplier addresses the limitations of generic financial AI tools by deploying repeatable infrastructure directly within a customer's existing systems, customized to match each firm's specific data, workflows, and security requirements. To prevent performance degradation and bias during continuous learning, Multiplier builds custom evaluations tailored to each client's investment framework, backed by human oversight. Early adopters report that Multiplier streamlines workflows across investment processes, ranging from initial idea generation to post-trade reviews, by organizing internal research and historical decisions. Early customer Mercator Partners, which helped incubate the firm, noted a significant shift in time spent from gathering information to executing decisions. Multiplier plans to use the newly raised funds to expand its infrastructure and develop its platform into a central operating system for asset management firms.

1

Rep. Ritchie Torres Moves to Ban Insider Trading in Political Prediction Markets Representative Ritchie Torres is taking a proactive stance against potential abuses in the burgeoning field of political prediction markets. Prompted by a controversial $400,000 profit made from a bet on Venezuelan President Nicolás Maduro’s capture, Torres plans to introduce the Public Integrity in Financial Prediction Markets Act of 2026. This legislation aims to prohibit federal officials and employees with access to nonpublic political information from trading on platforms that predict government actions or political outcomes, aligning these markets with existing insider trading rules in traditional finance. The impetus for this bill was a suspiciously timed large wager on Polymarket that predicted Maduro’s removal from power. Shortly after, US forces reportedly captured Maduro, leading to a rapid payout to the trader who made the bet from an account with little prior activity. This raised concerns about whether insider political or military intelligence was exploited. Polymarket has since faced scrutiny, including security breaches attributed to a third-party authentication flaw, leading to drained user balances and a crackdown on account security. Torres’ push highlights a crucial intersection of emerging financial technologies and regulatory frameworks, emphasizing the need for transparency and fairness as prediction markets grow in popularity. The bill will likely set precedents for applying insider trading principles to new digital trading forms tied to political events. As prediction markets evolve, this move underscores the importance of safeguarding market integrity against misuse of privileged information. #InsiderTrading #PredictionMarkets #PoliticalRegulation

1

In Q3, payment applications on Polygon Labs processed $1.82B in transfer volume, up 49% QoQ across 50+ platforms, according to Messari’s State of Polygon Q3 2025. Within that momentum, BlindPay (YC W25) reached $198.7M, delivering 91.7% QoQ growth and ranking among the top four payment apps by Q3 transfer volume. Stablecoin usage in payments continues to rise, and Polygon’s low fees and fast finality make it a natural home for these flows. BlindPay’s infrastructure delivers an enterprise-grade experience that’s simple to integrate, easy to scale, ready for global reach, and backed by compliance built in from day one. We are proud to build alongside other ecosystem contributors helping to scale payments on Polygon, including Paxos, Avenia, Revolut, and Cobo. 

90

15 Comments

The future of payments is being built on crypto rails. During his talk at the Blockworks Digital Asset Summit 2025, Algorand CSMO Marc Vanlerberghe highlighted the Pera Wallet spending card as an example of how that future is taking shape.

69

1 Comment

Circle's Q4 numbers show USDC gaining ground as regulated stablecoin infrastructure matures. For LATAM: the institutional shift from USDT to compliant alternatives is accelerating. The winning layer won't be the coin it'll be the compliance + ops middleware that makes corridors bankable.

5

1 Comment