Making Open Source Ai Win 🩷
Remote
Joined May 2023
- Tweets10.1K
- Following2.6K
- Followers4.4K
- Likes24.1K
2/2 | Solver + Evolver: learning from actual failures
The Solver attempts every task three times in isolated containers. Repeated attempts help distinguish a persistent failure pattern from a one-off mistake.
The Evolver is a stateful Claude Code session with bash access.
It examines the Solver's transcripts and working files, runs scripts to investigate why attempts failed, tests its hypotheses, then rewrites the agent's skills for the next round.
There is an important evaluation boundary: the Evolver can use training feedback, but it does not see validation or held-out scores.
That gives builders a repeatable cycle:
Execute → inspect failures → test explanations → revise skills → evaluate.
The model stays fixed. The instructions evolve through evidence gathered from the agent's own work.
21
1/2 | Benchmark Adaptor: removing the setup bottleneck
A skill optimizer is only useful if it can work with the benchmark you care about.
EvoSkill v2 addresses the setup friction with its Benchmark Adaptor.
You provide:
- A plain-language description of the benchmark
- The task data and split
- A grader
The Adaptor converts these into a frozen, hashed benchmark plug-in and a read-only reference folder.
That means builders don't have to manually translate every custom benchmark into an optimizer-specific format before they can start evolving skills.
The evaluation setup is established before the optimization loop begins.
1
1
25
Agent improvement has relied heavily on manual prompt iteration, hand-written skills and expensive model-level changes.
EvoSkill v2 introduces a different workflow: an automated, multi-agent pipeline for evolving the instructions around an agent while keeping its underlying model and runtime unchanged.
Benchmark setup, repeated task execution, failure analysis and skill revision become parts of one structured loop.
The important engineering detail is that the system can learn from execution traces and turn recurring failures into reusable instructions.
You can improve the agent's working process without retraining its model.
@SentientAGI
@sentient_found
@SentientEco
@abhishek095
EvoSkill v2 uses a three-agent system to improve your own agent without touching the model:
1) Benchmark adaptor builds a plugin from your grader, data, and plain-language description of the benchmark.
2) Solver attempts each task three times.
3) Evolver reads attempts and rewrites skills each round.
2
3
89
Agent improvement has relied heavily on manual prompt iteration, hand-written skills and expensive model-level changes.
EvoSkill v2 introduces a different workflow: an automated, multi-agent pipeline for evolving the instructions around an agent while keeping its underlying model and runtime unchanged.
Benchmark setup, repeated task execution, failure analysis and skill revision become parts of one structured loop.
The important engineering detail is that the system can learn from execution traces and turn recurring failures into reusable instructions.
You can improve the agent's working process without retraining its model.
@SentientAGI
@sentient_found
@SentientEco
@abhishek095
EvoSkill v2 uses a three-agent system to improve your own agent without touching the model:
1) Benchmark adaptor builds a plugin from your grader, data, and plain-language description of the benchmark.
2) Solver attempts each task three times.
3) Evolver reads attempts and rewrites skills each round.
3
5
293
RANA retweeted
The biggest challenge in open-source AI is not producing frontier research, it is translating that research into reliable, accessible primitives that developers can actually run in production.
Bringing in Julie Beliao (ex-Mozilla.ai, Google, PhD in NLP) as VP of Product signals exactly where Sentient is heading: turning breakthroughs like EvoSkill and model provenance benchmarks into standardized developer infrastructure.
Huge milestone for the ecosystem. Welcome to @SentientAGI! 🩷
The biggest challenge in open-source AI is not producing frontier research, it is translating that research into reliable, accessible primitives that developers can actually run in production.
Bringing in Julie Beliao (ex-Mozilla.ai, Google, PhD in NLP) as VP of Product signals exactly where Sentient is heading: turning breakthroughs like EvoSkill and model provenance benchmarks into standardized developer infrastructure.
Huge milestone for the ecosystem. Welcome to @SentientAGI! 🩷
RANA retweeted
Two numbers in this week’s OS AI Field Notes frame the entire market right now:
1. Anthropic’s S-1 showed ~85% gross margins on closed models.
2. The @ArtificialAnlys Progress Index shows the gap between open-weight models and the proprietary frontier has compressed to just 2 months (led by Kimi K3.1 Pro).
Rented API moats don't hold when open models catch up 60 days later at 20x lower cost.
• On-Device Scaling: @liquidai introduced LFM3, outperforming Gemini 3.5 Flash Lite on hardware efficiency.
• Enterprise Reality: @modal CEO Erik Bernhardsson made clear on 20VC why open-source inference is accelerating enterprises need custom, private deployments, not generic black boxes.
• Agent Evolution: Following Sentient’s EvoSkill v2 paper (beating Microsoft’s SkillOpt on held-out benchmarks), @sentient_found is co-hosting Open AGI Builder’s Day with Shanghai Jiao Tong University (@sjtu1896) on Nov 15.
Read the full November 6 Field Notes: sentient.foundation/news/os-…
@SentientAGI 🤌
RANA retweeted
Anthropic’s S-1 revealed ~85% gross margins, but the moat is shrinking in real time.
According to the new @ArtificialAnlys Progress Index in this week's Field Notes, open-weight models historically trailed closed labs by 4 months and the newest release, Kimi K3.1 Pro, sits just 2 months behind the closed frontier at a fraction of the cost.
When closed APIs charge premium rents for a 60-day lead, open-source commoditization is inevitable.
@SentientAGI
🧵👇
Kimi K3.1 Pro ranks just behind Opus 5.5 & GPT-6 Astra at a lower cost, @liquidai ships a new model built for on-device AI, and @ArtificialAnlys finds open weights are only 2 months behind closed models.
Two numbers in this week’s OS AI Field Notes frame the entire market right now:
1. Anthropic’s S-1 showed ~85% gross margins on closed models.
2. The @ArtificialAnlys Progress Index shows the gap between open-weight models and the proprietary frontier has compressed to just 2 months (led by Kimi K3.1 Pro).
Rented API moats don't hold when open models catch up 60 days later at 20x lower cost.
• On-Device Scaling: @liquidai introduced LFM3, outperforming Gemini 3.5 Flash Lite on hardware efficiency.
• Enterprise Reality: @modal CEO Erik Bernhardsson made clear on 20VC why open-source inference is accelerating enterprises need custom, private deployments, not generic black boxes.
• Agent Evolution: Following Sentient’s EvoSkill v2 paper (beating Microsoft’s SkillOpt on held-out benchmarks), @sentient_found is co-hosting Open AGI Builder’s Day with Shanghai Jiao Tong University (@sjtu1896) on Nov 15.
Read the full November 6 Field Notes: sentient.foundation/news/os-…
@SentientAGI 🤌
This is why open AGI infrastructure matters.
Fresh off releasing EvoSkill v2 (which beat Microsoft and Alibaba on held-out agent benchmarks at $0.04/pt gained), @sentient_found is bringing builders together in person:
• Nov 14: Open AGI Tea House
• Nov 15: Open AGI Builder’s Day with Shanghai Jiao Tong University (@sjtu1896)
If you're building open, sovereign agent primitives, check out the Field Notes and apply to Sentient’s $42M Grant track:
sentient.foundation/news/os-…
1
32
Look at the infrastructure shipping across this edition:
• The Frontier Gap: Kimi K3.1 Pro is matching Opus 5.5 on CAD, game dev, and heavy STEM simulations.
• Edge Efficiency: @liquidai’s LFM3 is beating Gemini 3.5 Flash Lite on-device at lower resource overhead.
• Enterprise Sovereignty: As @modal CEO Erik Bernhardsson noted on 20VC, inference hasn't peaked enterprises are choosing open weights because workflow customization and data ownership beat rented API wrappers.
1
1
63
Anthropic’s S-1 revealed ~85% gross margins, but the moat is shrinking in real time.
According to the new @ArtificialAnlys Progress Index in this week's Field Notes, open-weight models historically trailed closed labs by 4 months and the newest release, Kimi K3.1 Pro, sits just 2 months behind the closed frontier at a fraction of the cost.
When closed APIs charge premium rents for a 60-day lead, open-source commoditization is inevitable.
@SentientAGI
🧵👇
Kimi K3.1 Pro ranks just behind Opus 5.5 & GPT-6 Astra at a lower cost, @liquidai ships a new model built for on-device AI, and @ArtificialAnlys finds open weights are only 2 months behind closed models.
RANA retweeted
Anthropic's S-1 reportedly shows gross margins of roughly 85%.
Meanwhile, the latest Artificial Analysis data puts the open-weight frontier just two months behind proprietary models.
That gap matters.
Kimi K3.1 Pro is closing in on frontier performance across coding, 3D modelling and financial research, while open models keep getting cheaper to run and easier to customize.
The pressure on closed AI economics is becoming harder to ignore.
When capable models can be deployed, adapted and run on your own infrastructure, enterprises gain options that a closed API cannot always provide.
Open source is becoming a serious economic force in AI.
@SentientAGI
